1 Stimmen dafür0 Stimmen dagegen

101 Aufrufe162 Seitentut qs

Jul 25, 2017

© © All Rights Reserved

PDF, TXT oder online auf Scribd lesen

tut qs

© All Rights Reserved

Als PDF, TXT **herunterladen** oder online auf Scribd lesen

101 Aufrufe

tut qs

© All Rights Reserved

Als PDF, TXT **herunterladen** oder online auf Scribd lesen

- arfima
- Insurance Risk and Ruin
- Probability Exercises
- R Hussain_1a
- Unsupervised Learning of Object Deformation Models
- 1-D Spectral Contour Coding For Content-Based Medical Image Retrieval
- Chap 7
- fifa
- Review of Use of Nonlocal Spectral – Spatial Structured Sparse Representation for Hyper spectral Imagery Restoration
- Data Mining
- Power-Law Process (PLP) (Gau Yan Xie)
- Test of Goodness of Fit
- 2008 Straub ADK Seismic Fragility
- Chapter 5
- Download
- BLIND SPEECH SEPARATION OF MOVING SPEAKERS IN REAL REVERBERANT ENVIRONMENTS
- lecture5 (1)
- Chap12
- Logit to Probit to LPM Example
- CDA OddsRatios 20160121

Sie sind auf Seite 1von 162

Exercises

Contents

1 Introduction 2

1.1 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2.1 Compound distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.2 Explicit claims count distributions . . . . . . . . . . . . . . . . . . . . . . . . . . 8

2.3 Parameter estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2.4 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

3.1 Data analysis and descriptive statistics . . . . . . . . . . . . . . . . . . . . . . . 32

3.2 Selected parametric claims size distributions . . . . . . . . . . . . . . . . . . . . 33

3.3 Model selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3.4 Calculating within layers for claim sizes . . . . . . . . . . . . . . . . . . . . . . . 36

3.5 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

4.1 Approximations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

4.2 Algorithms for compound distributions . . . . . . . . . . . . . . . . . . . . . . . 59

4.3 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

5.1 Solvency considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

5.2 Ruin theory in discrete time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

5.3 Ruin theory in continuous time . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

5.4 Dependence modelling and copulas . . . . . . . . . . . . . . . . . . . . . . . . . 81

5.5 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

6.1 Premium Risk-based Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

i

ACTL3162 General Insurance Techniques Exercises

6.3 Gaussian approximation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

6.4 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

7.1 Components of a GLM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

7.2 Deviance and Scaled Deviance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

7.3 Fit a GLM and Evaluate the quality of a model . . . . . . . . . . . . . . . . . . 110

7.4 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

8.1 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

8.2 Exact Bayesian models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

8.3 Linear credibility estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

8.4 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

9.1 Outstanding loss liabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

9.2 Claims reserving algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

9.3 Stochastic claims reserving methods . . . . . . . . . . . . . . . . . . . . . . . . . 144

9.4 Claims development result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

9.5 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

10.1 Decision theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

10.2 Game theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

10.3 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

1

Module 1

Introduction

1.1 Preliminaries

Exercise 1.1: [NLI1, Solution][?, Exercise 1]

P P P

(c) Assume X N (0, 1). Prove that E[X 2k+1 ] = 0 for all k N0 (natural numbers with

zero).

Exercise 1.2: [NLI2, Solution][?, Exercise 2] Assume that Xk has a 2 -distribution with k N

degrees of freedom, i.e. Xk is absolutely continuous with density

1

f (x) = k/2

xk/21 exp(x/2), for x 0. (1.1)

2 (k/2)

(a) Prove that f is a density (hint: see Section 3.3.3 and proof of Proposition 2.20 in ?).

(b) Prove

MXk (r) = (1 2r)k/2 for r < 1/2. (1.2)

d

(c) Choose Z N (0, 1) and prove Z 2 = X1 .

i.i.d Pk d

(d) Choose Z1 , ..., Zk N (0, 1). Prove i=1 Zi2 = Xk and calculate the first two moments of

the latter.

1.2 Solutions

Solution 1.1: [NLI1, Exercise]

2 b2 /2

Ma+bX (r) = E[er(a+bX) ] = era E[erbX ] = era MX (rb) = era+r , (1.3)

2

ACTL3162 General Insurance Techniques Exercises

P

(b) We can write down the moment generating function for i Xi (using the assumption of

independence),

!

Y Y X X

MPi Xi (r) = MXi (r) = exp(ri + r2 i2 /2) = exp r i + r 2 i2 /2 , (1.4)

i i i i

P P

(c) It is tempted to use the moment generating function to solve this exercise, but obtaining

explicit formula for the n-th derivative of the m.g.f. is not so easy. We will resort back

to the old fashion way. First denote E[X 2k+1 ] = I2k+1 for k N0 , then

Z

2k+1 1 1 2

I2k+1 = x exp x dx

2 2

Z 1 2

1 d exp x

= x2k 2

dx

2 dx

x= Z

1 2k 1 2 1 2k1 1 2

= x exp x + 2k x exp x dx

2 2 x= 2 2

1

= 2k I2k1 . (1.5)

2

(2k2)

This recursive formula keeps on going and yields I2k+1 = 2k . . . 22 I1 . Since I1 =

2 2

E[X] = 0, therefore we have E[X 2k+1 ] = 0 for all k N0 .

(a) To show that f is a density function, first note that f (x) 0 for all x 0. Secondly

using a gamma density with shape parameter k/2 for k N and scale parameter 1/2, we

have

Z

1

k/2

xk/21 exp(x/2)dx = 1. (1.6)

0 2 (k/2)

Z

1

MXk (r) = exp(rx) k/2 xk/21 exp(x/2)dx

0 2 (k/2)

Z

1

= k/2 xk/21 exp((1/2 r)x)dx

2 (k/2) 0

(1/2 r)k/2 (k/2)

= , if 1/2 r > 0

2k/2 (k/2)

= (1 2r)k/2 . (1.7)

F (x) = P(Z 2 x) = P( x Z x)

Z x

1 1 2

= exp z dz

x 2 2

Z x

1 1 2

=2 exp z dz. (1.8)

0 2 2

3

ACTL3162 General Insurance Techniques Exercises

Let z= y 1/2 and dz = 1/2y 1/2 dy, using this change of variable, we have z changes from

0 to x and y changes from 0 to x. The cumulative probability function F (x) becomes

Z x

1 1

F (x) = exp y y 1/2 dy

2 2

Z0 x

1 1

= 1/2 (1/2)

exp y y 1/2 dy, (1.9)

0 2 2

i.i.d.

(d) Suppose Z1 , ..., Zk N (0, 1), then Z12 , ..., Zk2 are i.i.d. Chi-square random variables with

2 d

Pk

1 degree of freedom. Then to prove i=1 Zi = Xk , it suffices to show that the sum

Pk i 1 2 k

i=1 X1 has Chi-square distribution with k degrees of freedom, where X1 , X1 , ..., X1 are

i.i.d. Chi-square random variables with 1 degree of freedom. This can be done using the

moment generating function,

k

Y k

Y

MPk Zi2 (r) = MX1i (r) = (1 2r)1/2 = (1 2r)k/2 . (1.10)

i=1

i=1 i=1

4

Module 2

Exercise 2.1: [los3K, Solution] [?, Problem 2.2.2] Throw a true die and let X denote the

outcome. Then, toss a coin X times. Let Y denote the number of heads obtained. What are

the expected value and the variance of Y ?

Exercise 2.2: [los7, Solution] Let the sum S = X1 + X2 + X3 with X1 , X2 and X3 distributed

as follows:

x f1 (x) f2 (x) f3 (x)

0 0.2 - 0.5

1 0.3 - 0.5

2 0.5 p -

3 - 1p -

4 - - -

where 0 < p < 1. You are also given that FS (4) = 0.43. Calculate the value of p.

0 0.1 0.0 0.30

1 0.2 0.2 0.10

2 0.2 0.0 0.05

3 0.2 0.3 0.30

4 0.2 0.4 0.15

5 0.1 0.1 0.10

Calculate

1. The distribution of X1 + X2

2. The distribution of X1 + X2 + X3

5

ACTL3162 General Insurance Techniques Exercises

Exercise 2.4: [los11, Solution] An insurance portfolio has the following characteristics:

Distribution of N Distribution of X

n P (N = n) x p (x)

0 0.4 1 0.6

1 0.3 2 0.4

2 0.3

Number of claims, N , and claim amounts, X, are mutually independent. Compute the condi-

tional probability that the average claim size exceeds the expected claim size, given that there

is at least one claim.

S = X 1 + X2 + + XN

where:

n P (N = n)

0 0.3

1 0.2

2 0.5

Exercise 2.6: [los13K, Solution] [?, Problem 2.3.1] Calculate Pr[S = s] for s = 0, 1, . . . , 6

when S = X1 + 2X2 + 3X3 and Xj Poisson(j).

in function of i and fXi (x), i = 1, 2, 3, if the i s are positive integers and for any non-

negative range of Xi , i = 1, 2, 3. You should not have to change your program if the

distributions or the i s change.

To check your program, try to run it with 1 = 3, 2 = 0 and 3 = 2, as well as the

same probabilities as in Exercise 2.3, except for fX1 (5) = fX1 (6) = 0.05. Here is the

distribution you should get:

x Solution

1 0 0.0300

2 1 0.0000

3 2 0.0100

4 3 0.0600

5 4 0.0050

6 5 0.0200

6

ACTL3162 General Insurance Techniques Exercises

7 6 0.0900

8 7 0.0100

9 8 0.0350

10 9 0.1200

11 10 0.0200

12 11 0.0500

13 12 0.1200

14 13 0.0300

15 14 0.0500

16 15 0.0750

17 16 0.0300

18 17 0.0350

19 18 0.0750

20 19 0.0225

21 20 0.0350

22 21 0.0150

23 22 0.0225

24 23 0.0075

25 24 0.0150

26 25 0.0050

27 26 0.0075

28 27 0.0000

29 28 0.0050

Exercise 2.8: [los21R, Solution][R ] Consider Exercise 2.6 above. Can you use the R code

developed in Exercise 2.7 to get the distribution of S? Why?

X

fS (x) = pn (x) Pr[N = n]

n=0

in R as a function of the distributions of N and X. What are the conditions for this to be

feasible? Print the pmf and df of S, as well as the descriptive statistics of S. Try to make your

code as efficient as possible.

Check your code with ?, Example 12.2.2.

Use Theorem 12.4.2 to develop a program in R that calculates and displays fS (x) and FS (x)

for x s, in function of the inputs s, and i , i = 1, . . . , m.

Check your results with Exercise 1.12 and plot fS (x) for x 36. You should get:

x f_S F_S

1 0 0.0024787522 0.002478752

2 1 0.0024787522 0.004957504

3 2 0.0061968804 0.011154385

4 3 0.0128068862 0.023961271

5 4 0.0149757944 0.038937065

6 5 0.0243950527 0.063332118

7 6 0.0332600344 0.096592153

[...]

37 36 0.0004111363 0.999152382

7

ACTL3162 General Insurance Techniques Exercises

0.06

0.05

0.04

results$f_S

0.03

0.02

0.01

0.00

0 5 10 15 20 25 30 35

results$x

Exercise 2.11: [los10, Solution] Calculate E[S], V ar(S) and mS (t) in case N has the following

distribution:

1. Poisson()

2. binomial(n, p)

3. negative binomial(r, p)

Exercise 2.12: [NLI2.7, Solution][?, Corollary 2.7] Assume S1 , ..., Sn are independent with

Sj CompBinom(vj , p, G) for all j = 1, ..., n. Show that the aggregated claim has a compound

binomial distribution with

n

X Xn

S= Si CompBinom( vj , p, G). (2.1)

i=1 j=1

1, 2, 3, 4. Calculate Pr[S = s] for s 4:

Exercise 2.14: [los15K, Solution] [?, Problem 3.5.8] Assume that S1 is compound Poisson

distributed with parameter = 2 and claim sizes p(1) = p(3) = 21 . Let S2 = S1 + N , where

N is Poisson(1) distributed and independent of S1 . Determine the mgf of S2 . What is the

corresponding distribution? Determine Pr[S2 2.4]. Leave the powers of e unevaluated.

8

ACTL3162 General Insurance Techniques Exercises

Exercise 2.15: [los16, Solution] You are given S = S1 + S2 , where S1 and S2 are independent

and have compound Poisson distributions with 1 = 3 and 2 = 2 and individual claim amount

distributions:

x p1 (x) p2 (x)

1 0.25 0.10

2 0.75 0.40

3 0.00 0.40

4 0.00 0.10

Determine the mean and variance of the individual claim amount for S.

Exercise 2.16: [los17K, Solution] [?, Problem 3.4.3] Assume that S1 is compound Poisson

with 1 = 4 and claims p1 (j) = 14 , j = 0, 1, 2, 3, and S2 is also compound Poisson with 2 = 2

and p2 (j) = 21 , j = 2, 4. If S1 and S2 are independent, then what is the distribution of S1 + S2 ?

Exercise 2.17: [los18, Solution] Suppose that S has a compound Poisson distribution with

parameter and with discrete claims distribution

S= xi Ni

i=1

where Ni denotes the frequency of the claim amount xi for i = 1, 2, ..., m.

2. Prove that each Ni has a Poisson distribution and find its parameter.

Exercise 2.18: [los19, Solution] Suppose that the number of accidents incurred by an insured

driver in a single year has a Poisson distribution with parameter . If an accident happens, the

probability is p that the damage amount will exceed a deductible (or excess) amount. On the

assumption that the number of accidents is independent of the severity of the accidents, derive

the distribution of the number of accidents that result in a claim payment.

Exercise 2.19: [sur1K, Solution] [?, Problem 4.2.2] Let {N (t), t 0} be a Poisson process

with parameter , and let pn (t) = Pr[N (t) = n] and p1 (t) 0. Show that

[Express pn (t + dt) as a function of pn (t) and interpret this relation.]

Exercise 2.20: [NLI6, Solution][?, Exercise 6] An insurance company decides to offer a no-

claims bonus to good car drivers, namely,

9

ACTL3162 General Insurance Techniques Exercises

How does the base premium need to be adjusted so that this no-claims bonus can be financed?

For simplicity we assume that all risks have been insured for at least 6 years. Answer the

question in the following two situations:

(a) Homogeneous portfolio with i.i.d. risks having i.i.d. Poisson claim counts with frequency

parameter = 0.2.

(b) Heterogeneous portfolio with independent risks being characterised by a frequency param-

eter having a gamma distribution with mean = 0.2 and Vco() = 1 (Vco stands for

coefficient of variation). Conditionally, given , the individual years have i.i.d. Poisson

claim counts with frequency parameter .

Exercise 2.21: [Fit1, Solution] Consider a set of n right-censored insurance claims observations

where the observed data is represented as

where xj refers to the claim amount observed and j is the (right-censor) indicator whether

the applicable policy limit has been reached. You are to fit a (simple) exponential distribution

model to the observed claims with

1. Write down the log-likelihood function for estimating the exponential parameter.

3. Describe how you can derive a standard error of your parameter estimate.

Exercise 2.22: [Fit2, Solution] [Institute of Actuaries, Subject 106, September 2000] An insur-

ance company has a portfolio of policies with a per-risk excess of loss reinsurance arrangement

with a deductible of M > 0. Claims made to the direct insurer, denoted by X, have a Pareto

distribution with cumulative distribution function

200

FX (x; ) = 1 .

200 + x

There were a total of n claims from the portfolio. Of these claims, I were for amounts less than

the deductible. The claims less than the deductible are

xi , for i = 1, 2, ..., I.

PI

The value of the statistic i=1 log (200 + xi ) = y is given.

y.

10

ACTL3162 General Insurance Techniques Exercises

b = 1.75.

(b) Assuming that = 1.75, estimate the average amounts paid by the insurer and the

reinsurer on a claim made during the year.

Exercise 2.23: [Fit3, Solution] Observations (which are number of claims) Y1 , Y2 , ..., Yn are

independent Poisson random variables with E(Yi ) = i where

, for i = 1, 2, ..., m

log i =

+ , for i = m + 1, m + 2, ..., n

Exercise 2.24: [Fit4, Solution][?, Exercise 12.52] Consider the Inverse Gaussian distribution

with density function expressed as

1/2 " 2 #

x

fX (x) = exp , for x > 0.

2x3 2x

1. Show that n n

(xj )2

X X 1 1 n

= 2

+ (x )2 ,

j=1

xj j=1

xj x x

Pn

where x = (1/n) j=1 xj .

2. For a given sample x1 , x2 , ..., xn , show that the maximum likelihood estimates of and

are

b=x

and

n

b = .

Pn 1 1

j=1

xj x

Exercise 2.25: [Fit5, Solution] The following 20 claim amounts were observed over a period

of time:

132 149 476 147 135 110 176 107 147 165

135 117 110 111 226 108 102 108 227 102

You are interested in estimating the probability that a claim will exceed 200. You are to fit the

Pareto distribution with cumulative distribution function of the form

2. Use this to estimate the probability that a claim will exceed 200.

Exercise 2.26: [NLI8, Solution][?, Exercise 8] Natural hazards in Switzerland are covered by

the so-called Schweizersische Elementarschade-Pool (ES-Pool). This is a pool of private Swiss

insurance companies which organises the diversification of natural hazards in Switzerland.

For pricing of these natural hazards one distinguishes between small events and large events,

the latter having a total claim amount exceeding CHF 50 millions per events. The following

11

ACTL3162 General Insurance Techniques Exercises

15 storm and flood events have been observed in years 1986-2005 (these are the events with a

total claim amount exceeding CHF 50 millions).

20.06.1986 52.8 18.05.1994 78.5

18.08.1986 135.2 18.02.1999 75.3

18.07.1987 55.9 12.05.1999 178.3

23.08.1987 138.6 26.12.1999 182.8

26.02.1990 122.9 04.07.2000 54.4

21.08.1992 55.8 13.10.2000 365.3

24.09.1993 368.2 20.08.2005 1051.1

08.10.1993 83.8

Fit a Pareto distribution with parameters = 50 and > 0 to the observed claim sizes.

Estimate parameter using the unbiased version of the MLE.

We introduce a maximal claims cover of M = 2 billions CHF per event, i.e. the individual

claims are given by Yi M = min{Yi , M } (see also Section 3.4.2 in ?). For the yearly

claim amount of storm and flood events, we assume a compound Poisson distribution

with Pareto claim sizes Yi . What is the expected total yearly claim amount?

What is the probability that we observe a storm and flood event next year which exceeds

the level of M = 2 billions CHF?

2.4 Solutions

Solution 2.1: [los3K, Exercise] Firstly, one should notice that X is a discrete uniform random

variable over {1, 2, 3, 4, 5, 6}. Therefore we have

1+2+3+4+5+6 7

E(X) = = ,

6 2

2

2 2 12 + 22 + 32 + 42 + 52 + 62 7 35

V ar(X) = E(X ) (E(X)) = = .

6 2 12

We have Y |X = x binomial(x, 1/2). Hence,

E (Y ) = E [E (Y |X )] = E [1/2X] = 7/4

and

V ar (Y ) = V ar [E (Y |X )] + E [V ar (Y |X )]

= V ar [1/2X] + E [1/4X] = 77/48.

x f1 (x) f2 (x) f1+2 (x) f3 (x) f1+2+3 (x) FS (x)

0 0.2 0 0 0.5 0 0

1 0.3 0 0 0.5 0 0

2 0.5 p 0.2p 0 0.1p 0.1p

3 0 1 p 0.2 + 0.1p 0 0.1 + 0.15p 0.1 + 0.25p

4 0 0 0.3 + 0.2p 0 0.25 + 0.15p 0.35 + 0.4p

12

ACTL3162 General Insurance Techniques Exercises

x fX1 (x) fX2 (x) fX1 +X2 (x)

0 0.1 0.0 0.00 = 0.1 0

1 0.2 0.2 0.02 = 0.1 0.2 + 0.2 0

2 0.2 0.0 0.04 = 0.1 0 + 0.22 + 0.2 0

3 0.2 0.3 0.07 = 0.1 0.3 + 0.2 0 + 0.22 + 0.2 0

4 0.2 0.4 0.14 = 0.1 0.4 + 0.2 0.3 + 0.2 0 + 0.22 + 0.2 0

5 0.1 0.1 0.19 = 0.12 + 0.2 0.4 + 0.2 0.3 + 0.2 0 + 0.22 + 0.1 0

6 0.18 = 0.2 0.1 + 0.2 0.4 + 0.2 0.3 + 0.2 0 + 0.1 0.2

7 0.16 = 0.2 0.1 + 0.2 0.4 + 0.2 0.3 + 0.1 0

8 0.13 = 0.2 0.1 + 0.2 0.4 + 0.1 0.3

9 0.06 = 0.2 0.1 + 0.1 0.4

10 0.01 = 0.12

P

Check your results by verifying that x fX1 +X2 (x) = 1.

2. In order to get the distribution of X1 + X2 + X3 , it suffices to calculate the convolution

of X1 + X2 with X3 . We have then

x fX3 (x) fX1 +X2 (x) fX1 +X2 +X3 (x)

0 0.30 0.00 0

1 0.10 0.02 0.006

2 0.05 0.04 0.014

3 0.30 0.07 0.026

4 0.15 0.14 0.057

5 0.10 0.19 0.0895

6 0.18 0.109

7 0.16 0.132

8 0.13 0.149

9 0.06 0.1355

10 0.01 0.1095

11 0.085

12 0.054

13 0.025

14 0.0075

15 0.001

Solution 2.4: [los11, Exercise] Let S = X1 + + XN denote the aggregate claims. Then we

wish to compute the probability P NS > E (X) |N > 0 where E (X) = 1.4. Thus, we have

S Pr [S > 1.4N N > 0]

Pr > E (X) |N > 0 =

N Pr [N > 0]

Pr [S > 1.4 |N = 1 ] Pr [N = 1] + Pr [S > 2.8 |N = 2 ] Pr [N = 2]

=

h i 1 0.4

(0.4) (0.3) + 1 (0.6)2 (0.3)

= = 0.52.

0.6

13

ACTL3162 General Insurance Techniques Exercises

X

FS (x) = P n (x) Pr[N = n]

n=0

X

P (S < 4) = P (X1 + + XN < 4 |N = n) P (N = n)

n=0

= 0.3 + P (X1 < 4) 0.2 + P (X1 + X2 < 4) 0.5

3

P (X1 < 4) =

4

and Z 3 Z 4x1

1 1 1

P (X1 + X2 < 4) = dx2 dx1 = .

1 1 4 4 8

Thus,

3 1

P (S < 4) = 0.3 + 0.2 + 0.5 = 0.5125.

4 8

e1 e2 e3 e3 e6

0 1 1 1 1 1

1 1 0 1 0 1

2 1/2 2 2 1/2 0 2 1/2

3 1/6 0 2 1/6 3 5 1/6

4 1/24 2 3 1/24 0 6 1/24

5 1/120 0 2 41/120 0 9 101/120

6 1/720 1 1/3 2 301/720 4 1/2 13 301/720

e1 /x! e2 2x/2 / (x/2)! e3 3x/3 / (x/3)!

1. Try to copy this code and paste this code in a R document. If you source it, you should

find the same results as in Exercise 1.8.1 and 1.8.2.

14

ACTL3162 General Insurance Techniques Exercises

fX1 <- c(.1,.2,.2,.2,.2,.1)

fX2 <- c(0,.2,0,.3,.4,.1)

fX3 <- c(.3,.1,.05,.3,.15,.1)

#####################

# convolution 1.8.1 #

#####################

fX12_8a <- c()

# this is to avoid NAs in the recursions afterwards

# the appropriate number depends on the range of the sum, here [0,10]

fX1 <- c(fX1,rep(0,5))

fX2 <- c(fX2,rep(0,5))

for(i in 1:11) # we know that the range of X_1+X_2 is [0,10]

{#now we calculate the probability

fX12_8a[i] <- sum( fX1[1:i]*fX2[i:1] )

}# end of the for loop

#print results

results<-data.frame(x=c(0:10),DistrX1=fX1,DistrX2=fX2,Solution=fX12_8a)

print("Exercise 1.8.1")

print(results)

#####################

# convolution 1.8.2 #

#####################

#use the same algorithm as above, but with fX12_8a and fX3...

fX123_8b <- c()

# add 0s

fX12_8a <- c(fX12_8a,rep(0,5))

fX3 <- c(fX3,rep(0,10))

for(i in 1:16) # we know that the range of X_1+X_2+X_3 is [0,15]

{#now we calculate the probability

fX123_8b[i] <- sum(fX3[1:i]*fX12_8a[i:1])

}# end of the for loop

#print results

resultsb<-data.frame(x=c(0:15),DistrX3=fX3,DistrX1X2=fX12_8a,Solution=fX123_8b)

print("Exercise 1.8.2")

print(resultsb)

x DistrX1 DistrX2 Solution

1 0 0.1 0.0 0.00

2 1 0.2 0.2 0.02

3 2 0.2 0.0 0.04

15

ACTL3162 General Insurance Techniques Exercises

5 4 0.2 0.4 0.14

6 5 0.1 0.1 0.19

7 6 0.0 0.0 0.18

8 7 0.0 0.0 0.16

9 8 0.0 0.0 0.13

10 9 0.0 0.0 0.06

11 10 0.0 0.0 0.01

[1] "Exercise 1.8.2"

x DistrX3 DistrX1X2 Solution

1 0 0.30 0.00 0.0000

2 1 0.10 0.02 0.0060

3 2 0.05 0.04 0.0140

4 3 0.30 0.07 0.0260

5 4 0.15 0.14 0.0570

6 5 0.10 0.19 0.0895

7 6 0.00 0.18 0.1090

8 7 0.00 0.16 0.1320

9 8 0.00 0.13 0.1490

10 9 0.00 0.06 0.1355

11 10 0.00 0.01 0.1095

12 11 0.00 0.00 0.0850

13 12 0.00 0.00 0.0540

14 13 0.00 0.00 0.0250

15 14 0.00 0.00 0.0075

16 15 0.00 0.00 0.0010

#############

# variables #

#############

# vectors of probabilities

fX1 <- c(.1,.2,.2,.2,.2,.05,.05)

fX2 <- c(0,.2,0,.3,.4,.1)

fX3 <- c(.3,.1,.05,.3,.15,.1)

# weights

alpha <- c(3,0,2)

################

# convolutions #

################

#just to be sure..

if(sum(alpha)==1) print("there is no convolution to do!")

Xmax <- c(length(fX1)-1,length(fX2)-1,length(fX3)-1)

rangeS <- sum(Xmax*alpha)

fX <- array(c(fX1,rep(0,rangeS-Xmax[1]),

fX2,rep(0,rangeS-Xmax[2]), #we complete each colun with 0s

fX3,rep(0,rangeS-Xmax[3])), #end of values

c(rangeS+1,3)) #now the dimension of the array (second argument)

#####

#note that we have at most only two convolutions to do

16

ACTL3162 General Insurance Techniques Exercises

#initialising results array - 3 columns for scaled Xs, one for convolution

#of first two, last for the solution

fS <- array(rep(0,(rangeS+1)*5),c(rangeS+1,5))

for(i in 1:3) { # for each X

if(alpha[i]==0) next

for(j in 1:(Xmax[i]+1)) {

fS[(j-1)*alpha[i]+1,i] <- fX[j,i]

} # end of j loop

}# end of i loop

for(i in 1:(rangeS+1)) {

if(alpha[2]==0){ # then only 1 and 3 need to be convoluted (see test above)

fS[i,5] <- sum(fS[1:i,1]*fS[i:1,3]) #end of the story

} else { if(alpha[1]==0){ #then only 2 and 3 need to be convoluted

fS[i,5] <- sum(fS[1:i,2]*fS[i:1,3]) #end of the story

} else { fS[i,4] <- sum(fS[1:i,1]*fS[i:1,2]) # we do 1 and 2 and see...

if(alpha[3]==0) { # if alpha 3 is 0, then it is finished

fS[i,5] <- fS[i,4] # ... and we translate results in column 5

# otherwise we do the last convolution 1*2 with 3:

} else {fS[i,5] <- sum(fS[1:i,3]*fS[i:1,4])}

} # end alpha 3 if

} #end second else

} # end for

#print results

results<-data.frame(x=c(0:rangeS),Solution=fS[,5])

print("Solution")

print(results)

plot(results)

#check we have a true distribution..:

if(min(fS[,5])<0) print("Oups some probabilities are negative") else

print("All probabilities are positive")

if(max(fS[,5])>1) print("Oups some probabilities are > 1") else

print("All probabilities are < 1")

print(c("The sum of them is: ",sum(fS[,5])))

17

ACTL3162 General Insurance Techniques Exercises

0.12

0.10

0.08

Solution

0.06

0.04

0.02

0.00

0 5 10 15 20 25

You can see the effect of the scaling as the pmf is not smooth.

Solution 2.8: [los21R, Exercise] The support of the Xi is not finite, which means that it is

not possible to calculate the exact distribution of S without truncating the distribution of Xi

at some points; otherwise the program will consider the support of S as infinite and there will

be infinite loops. However, it is possible to achieve a decent level of accuracy (if one is patient),

but that would require careful additional programming to determine where the distribution of

the Xi s should be truncated. This would also require analysis to check if the moments and

quantiles of S are reasonably conserved...

Note that it is not possible to develop a code for the general case

S = 1 X1 + 2 X2 + 3 X3 , i 0 (integers), i = 1, . . . , 3

if some of the Xi s have infinite support without truncating them. However, if all the Xi s are

Poisson, we can then use A Theorem 12.4.1 and allow for any number of i s.

Solution 2.9: [los22R, Exercise] First note that this formula can only be used if X and N are

both discrete and with a finite range. One possible code is as follows:

# inputs #

##########

fX <-c(0,.5,.4,.1)

fN <-c(.1,.3,.4,.2)

# program #

###########

# range of X, N and S

rX <- length(fX)-1

rN <- length(fN)-1

18

ACTL3162 General Insurance Techniques Exercises

rS <- rX*rN

#first create a vector with the contents of the array

cont <- c()

#first column

cont <- c(cont,1,rep(0,rS)) #p*0(x)

cont <- c(cont,fN[1]) #last line is distribution of N

#second column

cont <- c(cont,fX,rep(0,rS-rX),fN[2])

#following columns are 0s, except in the last line

for(i in 1:(rN-1)) { #there will be rN-1 convolutions

cont <- c(cont,rep(0,rS+1),fN[i+2])

} #end of the i loop

#last two columns for pmf and df

#(not in the loop to avoid n/as in the last row - a detail)

cont <- c(cont,rep(0,2*(rS+2)))

#and finally the array

distS <- array(cont,c(rS+2,rN+3))

if(fX[1]>0){

for(i in 1:(rN-1)) {

for(j in 1:((1+i)*rX+1)) { #lower side of triangle - see below

distS[j,i+2] <- sum(distS[1:j,2]*distS[j:1,i+1])

} # end j loop

} # end i loop

} else { #for efficiency: if f_X(0)=0, there will only be probabilities

# in a triangle with upper side with slope -1 from col 2

# and lower side with slope -rX (whether mass is >0 or not!)

for(i in 1:(rN-1)) {

for(j in 1:((1+i)*rX+1)) { #first change (lower side)

distS[j,i+2] <- sum(distS[1:max(j-i,1),2]*distS[(j+1)-(1:max(j-i,1)),i+1])

#second change (upper side) is max(j-1,1)

} # end j loop

} # end i loop

}

# first line (cant be put in the following loop)

distS[1,rN+2] <- sum(distS[1,1:(rN+1)]*distS[rS+2,1:(rN+1)]) #pmf at i

distS[1,rN+3] <- distS[1,rN+2]

# next lines

for(i in 2:(rS+1)) {

distS[i,rN+2] <- sum(distS[i,1:(rN+1)]*distS[rS+2,1:(rN+1)]) #pmf at i

distS[i,rN+3] <- distS[i-1,rN+3]+distS[i,rN+2]

} # end i loop

# print results #

#################

results <- data.frame(x=0:rS,pmf=distS[1:(rS+1),rN+2],df=distS[1:(rS+1),rN+3])

print(results)

source("path to program of Exercise 11.a")

pmf_to_desc_stats(distS[1:(rS+1),rN+2],1)

if(min(distS[1:(rS+1),rN+2])<0) print("Oups some probabilities are negative") else

print("All probabilities are positive")

if(max(distS[1:(rS+1),rN+2])>1) print("Oups some probabilities are > 1") else

print("All probabilities are < 1")

Efficiency comments:

1. Consider the convolutions part. In the bounds for j and for the convolutions, we recognise

that the probability masses of the convolutions will spread in the shape of a triangle

in the table of successive convolutions. The upper side of the triangle has slope 0 if

Pr[X = 0] > 0 and slope -1 otherwise. The lower side of the triangle will always have a

slope of minus the range of X. So you will have probabilities everywhere only if the range

of X is infinite (which will never happen in this program - see the note at the beginning of

the solution) and if Pr[X = 0] > 0. We can thus save resources by doing the convolutions

(products) only when these probabilities are different form zeros, which is achieved by

19

ACTL3162 General Insurance Techniques Exercises

2. Note the line

We use here the fact that R is a vector based program and we can perform j products in

a single line using 1:j and j:1 in the index of distS. This effectively does the same as

the following loop

temp <- 0

for(k in 1:j) {

temp <- temp + distS[k,2]*distS[j-k+1,i+1]

} # end k loop

distS[j,i+2] <- temp

which is a more traditional way to program it (in VBA or Maple or Mathematica, you

would program the convolution with the loop).

What impact do these considerations have? Here are the approximate processing time if

fX (x) = 0.01, x = 1, 2, . . . , 100 and fN (x) = 0.05, x = 0, 1, . . . , 19 (performed on an iMac

with 3.06 GHz Intel Core 2 Duo chip):

no triangle triangle

loop 2 min. 56 sec. 1 min. 5 sec.

vector 3 sec. 1.5 sec.

In this case, using the vector-based programming is much more important than being smart

about the triangle, but this is because we use R and we can take advantage of its very powerful

algorithms to handle vectors. In VBA, the triangle trick would be crucial.

Finally, note the shape of the pmf of S in this case (we omit here the mass at 0):

0.0012

0.0010

distS[2:(rS + 1), rN + 2]

0.0008

0.0006

0.0004

0.0002

0.0000

1:rS

20

ACTL3162 General Insurance Techniques Exercises

You can see the effect of compounding: when probabilities are due to only a few claims there

are irregularities (jumps) in the pmf (lhs of plot). When we are looking at outcomes of S that

involve a lot a different possible number of claims, the pmf is much smoother (rhs of plot).

Solution 2.10: [los23R, Exercise] Here is the code with the parameters corresponding to Ex-

ercise 1.12:

# Inputs #

##########

fX <- c(0,1/6,2/6,3/6)

l <- 6

s <- 36

# Program #

###########

lambda <- c()

alpha <- c()

for(i in 2:length(fX)) # we can discard alpha=0

{if(fX[i]>0) {

lambda <- c(lambda,fX[i]*l)

alpha <- c(alpha,i-1)}

} #end i loop

#number of variables

num <-length(alpha)

#number of columns that are necessary

#(the num distributions + num-1 convolutions)

colu <- 2*num -1

fS <- array(0,c(s+1,colu))

for(i in 1:num){

for(j in seq(1,(s+1),by=alpha[i])){ #we spread the probabilities

fS[j,i] <- exp(-lambda[i])*lambda[i]^((j-1)/alpha[i])/factorial(((j-1)/alpha[i]))

} # end of j loop

} # end of i loop

for(i in 1:(num-1)){

for(j in 1:(s+1)){

fS[j,num+i] <- sum(fS[1:j,num+i-1]*fS[j:1,num-i])

} # end j loop

} # end i loop

#calculate the df

FS <- c()

for(i in 1:(s+1)){FS[i]<-sum(fS[1:i,colu])}

#print results

results <- data.frame(x=c(0:s),f_S=fS[,colu],F_S=FS)

plot(results$x,results$f_S)

print(results)

E (N ) = ; V ar (N ) = ;

exp et 1 ;

mN (t) =

; V ar (S) = 2 + 2 ;

E (S) =

mS (t) = exp [ (mX (t) 1)] .

21

ACTL3162 General Insurance Techniques Exercises

Binomial:

E (N ) = np; V ar (N ) = npq;

n

mN (t) = q + pet ;

E (S) = np ; V ar (S) = np 2 + npq2 ;

mS (t) = [q + pmX (t)]n .

Negative Binomial:

E (N ) = r (1 p) /p; V ar (N ) = r (1 p) /p2 ;

r

p/ 1 (1 p) et ;

mN (t) =

E (S) = r (1 p) /p ; V ar (S) = r (1 p) /p 2 + r (1 p) /p2 2 ;

mS (t) = {p/ [1 (1 p) mX (t)]}r .

for all j = 1, ..., n. The moment generating function of Sj (see Proposition 2.6 from ?) for

j = 1, ..., n is

MSj (r) = (pMY1 (r) + (1 p))vj for r R. (2.2)

Since Sj for j = 1, ..., n are independent, then the moment generating function of the sum is

n

Y Pn

M j Sj (r) =

P

(pMY1 (r) + (1 p))vj = (pMY1 (r) + (1 p)) j=1 vj . (2.3)

j=1

Pn

Sj CompBinom( nj=1 vj , p, G).

P

This proves that j=1

0 1 0 0 0 0

1 0 0.1 0 0 0

2 0 0.2 0.01 0 0

3 0 0.3 0.04 0.001 0

4 0 0.4 0.10 0.006 0.0001

pn e2 2e2 2e2 4 2

3

e 2 2

3

e

Thus,

P (S e2 ;

= 0) =

P (S 0.2e2 ;

= 1) =

P (S 0.42e2 ;

= 2) =

P (S e2 (0.6 + 0.08 + 0.0013) = 0.6813e2 ;

= 3) =

2 4

P (S = 4) = e 0.8 + 0.2 + 0.006 + 0.000067 = 1.008067e2 .

3

22

ACTL3162 General Insurance Techniques Exercises

2. S compound Poisson( = 2) with p (x) = 0.1x, x = 1, 2, 3, 4. Write fj for the pdf with

Sj = 1 N1 + + j Nj , j = 1, 2, 3, 4, and pj for the pdf of j Nj so that

pj (x) = P (j Nj = x) = exp (0.2 j) (0.2 j)x/j / (x/j)!.

x p1 p2 = f2 p3 = f3 p4 = f4

0 .819 .670 .549 .549 .301 .449 .1353

1 .164 .110 .060 .0270

2 .016 .268 .231 .127 .0568

3 .001 .045 .329 .205 .0922

4 .000 .054 .048 .062 .359 .1364

N Poisson ( p(1) = 1) and

N3 Poisson ( p(3) = 1) ,

Now S2 = N1 + 3N3 with N1 being the sum of two independent N Poisson(1). Thus we have

N1 Poisson (1 = 1 + 1 = 2) and

N3 Poisson (3 = 1) again.

From this and Theorem 12.4.2, we observe that S2 is compound Poisson with = 3 and

1 3

P (X = 1) = = 2/3; P (X = 3) = = 1/3.

1 + 3 1 + 3

Next, invoke Panjer to compute

P (S 2) = P (S 2.4) .

Or using an ad hoc method:

P (S2 2.4) = P (N1 2 & N3 = 0)

= P (N1 2) P (N3 = 0)

= 20 /0! + 21 /1! + 22 /2! e2 e1 = 5e3 .

Solution 2.15: [los16, Exercise] From Theorem 12.4.1, we know S is compound Poisson with

= 5 and individual claims distribution

x P (x)

3

1 5

(0.25) + 25 (0.10) = 0.19

3

2 5

(0.75) + 25 (0.40) = 0.61

2

3 5

(0.40) = 0.16

2

4 5

(0.10) = 0.04

Thus, mean of individual claim amount is

1 0.19 + 2 0.61 + 3 0.16 + 4 0.04 = 2.05

and the variance of individual claim amount is

(1 2.05)2 0.19 + (2 2.05)2 0.61

+ (3 2.05)2 0.16 + (4 2.05)2 0.04

= 0.5075.

23

ACTL3162 General Insurance Techniques Exercises

and

4 1 1

p (0) = =

6 4 6

4 1 1

p (1) = =

6 4 6

4 1 2 1 2

p (2) = + =

6 4 6 2 6

4 1 1

p (3) = =

6 4 6

2 1 1

p (4) = = .

6 2 6

Solution 2.17: [los18, Exercise] Define the sum of the number of claims arising from each

possible claim amount by

Xm

N= Ni .

i=1

multinomial distribution with parameters n, 1 , 2 , ..., m . Its joint moment generating function

(m.g.f.) can be obtained as

" m

!#

" m

! #

X X X

E exp ti Ni = E exp ti Ni N = n P (N = n)

i=1 n=0 i=1

X n e n

= 1 et1 + 2 et2 + + m etm

n=0

n!

X n 1

= e 1 et1 + 2 et2 + + m etm

n=0

n!

m

!

X

= e exp i eti

i=1

Ym

exp i eti 1 .

=

i=1

ti = t and tj = 0 for all j 6= i, we obtain the m.g.f. of Ni which is given below:

Solution 2.18: [los19, Exercise] Denote N to be the number of accidents which is given to be

Poisson() distribution. Now, suppose N1 is the number of these accidents that lead to claims

(i.e. damage amount exceeds the deductible or excess). Then clearly, conditionally on N , N1

has a Binomial (N, p) distribution, since an accident leads to either a claim or no claim. Thus,

we have

m!

P (N1 = n |N = m) = pn (1 p)mn , n m.

n! (m n)!

24

ACTL3162 General Insurance Techniques Exercises

X

P (N1 = n) = P (N1 = n |N = m) P (N = m)

m=n

X m! e m

= pn (1 p)mn

m=n

n! (m n)! m!

e pn X 1

= m (1 p)mn

n! m=n (m n)!

e pn n X 1

= [ (1 p)]mn

n! m=n

(m n)!

e (p)n X 1

= [ (1 p)]k

n! k=0

k!

e (p)n (1p) ep (p)n

= e =

n! n!

which gives a Poisson distribution with parameter p.

This could have been anticipated (and also shown) as a consequence of Theorem 12.4.2.

= pn (t) + pn1 (t)

which gives the result for n = 0, 1, ... Note that for all n, we can write

Re-arranging, we get

Thus, the probability of having n jumps till t + dt is equal to the probability of getting n 1

jumps till t and another one within the next tiny interval dt, plus the probability of getting n

jumps till t and not other one within the next tiny interval dt.

Note that this expression could have been written directly using the law of total probability

and the property of the Poisson process seen in the lecture.

25

ACTL3162 General Insurance Techniques Exercises

(a) Assume that we are currently evaluating all the policies and we can break them down into

three categories: no claim for 6+ years (e60.2 = 0.3012), no claim for 3-6 years (e30.2

e60.2 = 0.2476174) and the rest. So about 30.12% of policies receive a 30% discount and

24.76% of policies receive a 10% discount. Now assume that the new premium is P and

we wish to solve

P 0.7 e60.2 + 0.9 (e30.2 e60.2 ) + 1 e30.2 = E[Y ] = P = 1.13 E[Y ] (2.4)

So in order to finance the bonus, we have the raise the base premium by 13 percent.

(b) For claims in the heterogeneous portfolio, we have assumed that the frequency parameter

follows a Gamma(, ) distribution. Using the fact that the mean = 0.2 and Vco() =

1, we can work out that = 1 and = 5. Thus, the probability of having zero claim in

one year (conditioning on values of ) is

Z

Prob[Number of Claims = 0|] = e 5 e5 d = 5/6. (2.5)

0

So the three categories have the following break-down, no claim for 6+ years ((5/6)6 =

0.334898), no claim for 3-6 years ((5/6)3 (5/6)6 = 0.2438057) and the rest. Using the

same method, the new premium we wish to solve is

P 0.7 (5/6)6 + 0.9 ((5/6)3 (5/6)6 ) + 1 (5/6)3 = E[Y ] = P = 1.142661 E[Y ],

(2.6)

which yields a 14% increase on the base premium.

Solution 2.21: [Fit1, Exercise] Note that there are no truncation in the observations and that

P (X > x) = ex .

1. The likelihood function for the observed data (xj , j ), j = 1, 2, ..., n can be written as

n

Y 1j j

L (; xj , j ) = exj exj

j=1

n

X

` (; xj , j ) = log L (; xj , j ) = [(1 j ) (log xj ) xj j ]

j=1

n

X

= [(1 j ) log xj + xj j xj j ]

j=1

n

X n

X

= log (1 j ) xj .

j=1 j=1

n n

` (; xj , j ) 1X X

= (1 j ) xj = 0

j=1 j=1

so that Pn

j=1 (1 j )

b = Pn .

j=1 xj

26

ACTL3162 General Insurance Techniques Exercises

3. Standard errors can be derived based on the second derivative of the log-likelihood. In

this case, it will be

n

2 ` (; xj , j ) 1 X

= 2 (1 j )

2 j=1

which should be negative at the optimum. So the standard error of our parameter estimate

will equal to

qP

#1/2 n

j=1 (1 j )

" n

1 X b

(1 j ) = P

q = P n .

j=1 xj

b2 j=1 n

(1 )

j=1 j

It is the square root of the negative of the inverse of the Hessian (which is the second

derivative) evaluated at the MLE.

Solution 2.22: [Fit2, Exercise] Note that the density of the Pareto can be expressed as

200

fX (x; ) = .

(200 + x)+1

Note also that the deductible of the Excess of Loss is equivalent to a policy limit from the point

of view of the insurer.

I (nI)

200

Y 200

L (; xi ) =

i=1

(200 + xi )+1 200 + M

I

I 200n Y

= (nI)

(200 + xi )1

(200 + M ) i=1

` (; xi ) = log L (; xi )

I

X

= I log + n log 200 (n I) log M ( + 1) log (200 + xi ) .

i=1

I

` (; xi ) I X

= + n log 200 (n I) log (200 + M ) log (200 + xi ) = 0

i=1

which gives

I

b= PI .

(n I) log (200 + M ) n log 200 + i=1 log (200 + xi )

You may wish to check that this gives the maximum by evaluating the second derivative:

2 ` (; xi ) I

2

= 2 < 0.

27

ACTL3162 General Insurance Techniques Exercises

400 400

b= = = 1.75.

100 log (800) 500 log 200 + 2209.269 228.57

Z

1.75 2001.75 200

E (X) = x 2.75 dx = = 266.67

0 (200 + x) 0.75

Z Z

(x M ) fX (x) dx = zfX (z + M ) dz

M 0

200 z800

Z

= dz

800 0 (800 + z)+1

1.75

1 800 1 800

= = = 94.28.

4 1 4 1.75 1

n

Y

L (, ; yi ) = ei (i )yi /yi !

i=1

so that log-likelihood is

n

X

` (, ; yi ) = log L (, ; yi ) = [i + yi log i log yi !]

i=1

m

X n

X

+

= [e + yi log yi !] + e + yi ( + ) log yi !

i=1 i=m+1

m

X n

X n

X

+

= me + yi (n m) e + ( + ) yi log yi !.

i=1 i=m+1 i=1

m n

` (, ; yi )

X

+

X

= me + yi (n m) e + yi = 0

i=1 i=m+1

and n

` (, ; yi ) X

= (n m) e+ + yi = 0.

i=m+1

These yield to

m

X n

X n

X

me + yi yi + yi = 0

i=1 i=m+1 i=m+1

28

ACTL3162 General Insurance Techniques Exercises

i=1 yi

b = log

m

and Pn Pm

i=m+1 yi i=1 yi

b = log log .

nm m

1. We have

n n

(xj )2 2

X X

= xj 2 +

j=1

xj j=1

xj

n n 2

2 2

X X

= + 2 + xj

j=1

xj x j=1

x

n

n2

2

X 1 1

= + 2n + nx

j=1

xj x x

n

2

X 1 1 n 2

2x + x2

= +

j=1

xj x x

n

X 1 1 n

= 2

+ (x )2

j=1

xj x x

"n

#

X (xj )2

L (, ; xj ) = c n/2 exp 2

2 j=1 xj

n

n X (xj )2

` (, ; xj ) = log c + log 2

2 2 j=1 xj

" n #

n X 1 1 n 2

= log c + log 2 2 + (x )

2 2 j=1

xj x x

" n 2 #

n X 1 1 n x

= log c + log + 1 .

2 2 j=1 xj x x

` 2n x x

= 1 2 =0

2 x

which implies

b=x

29

ACTL3162 General Insurance Techniques Exercises

" n 2 #

` n1 1 X 1 1 n x

= + 1 =0

2 2 j=1 xj x x

which implies

n

b = P .

n 1 1

j=1 xj

x

Solution 2.25: [Fit5, Exercise] Notice that the density of the given Pareto can be written as

fX (x) = 100 x1 .

Y

L () = 20 10020 x1

i

X

` () = 20 log + 20 log 100 ( + 1) log xi .

Differentiating, we get

` () 20 X

= + 20 log 100 log xi = 0

so that

20 20

b= P = = 2.848.

log xi 40 log 10 99.1252 92.1034

2. The required probability estimate is therefore

n

!1

1X

M LE = log Yi log . (2.7)

n i=1

n1

M LE = 0.9824864. (2.8)

n

30

ACTL3162 General Insurance Techniques Exercises

In previous part of the question, the fitted distribution for claim sizes Yi is a Pareto

distribution (50,0.9824864). Next we count the number of claims for each year and fit

them to a Poisson distribution. There are 2 claims in 1986, 2 in 1987, 1 in 1990, 1 in

1992, 2 in 1993, 1 in 1994, 3 in 1999, 2 in 2000 and 1 in 2005. These claim counts yield an

MLE estimator for the Poisson parameter of M LE = 5/3. The expected claim amount

(per claim) is

Z 2000 Z

E[min(Yi , 2000)] = yg(y)dy + 2000g(y)dy

0 2000

= I(G(2000)) + 2000(1 G(2000))

0.9824864+1 0.9824864

2000 2000

=1 + 2000

50 50

= 41.34829. (2.9)

So the expected total yearly claim amount is (using properties of compound Poisson),

5/3 41.34829 = 68.91382.

The probability that we observe a storm and flood event next year which exceeds the

level of 2 billions CHF is the product between probability of having one claim and claim

amount exceeds 2 billions,

Z

M LE

M LE e g(y)dy = 0.00648. (2.10)

2000

31

Module 3

Exercise 3.1: [los36R, Solution][R ] Develop a function that will yield a vector with the first

three central moments, 1 and 2 in function of the cgf of a random variable. One possible

beginning is

CMom123Gam12 <- function(cgf,param){

where cgf is an expression and where param is a list with the numerical values of the

parameters of cgf. The following commandfor an inverse Gaussian( = 2, = 4)

CMom123Gam12(expression(alpha*(1-sqrt(1-2*t/beta))),list(alpha=2,beta=4))

should yield

0.500000 0.125000 0.093750 2.121320 7.500000

1. Create a function that will calculate and return (to the assigned itemitem<-function())

a vector with E[], V ar(), 1 () and 2 () of a non-negative discrete random variable (with

finite range) in function of its pmf. In addition, a binary variable indicates if these de-

scriptive statistics should be printed in a data frame or not. Thus, the code (in a separate

R document) should look like

pmf_to_desc_stats <- function(pmf,print) {

[code omitted ^]

where pmf is a vector with the probabilities and where print is a binary (0-1) variable

indicating if the results should print or not.

2. Add this function to the code developed in Exercise 2.7 part 2 to print the descriptive

statistics. You should get something like

[1] "Descriptive statistics"

mean variance skewness kurtosis

1 12.05 35.1675 0.1213840 -0.4493699

32

ACTL3162 General Insurance Techniques Exercises

Exercise 3.3: [los1K, Solution] [?, Problem 3.8.1] Determine the mean and the variance of the

lognormal and the Pareto distribution. Proceed as follows: if Y lognormal(, 2 ), then ln Y

N (, 2 ); if Y Pareto(, x0 ), then Y /x0 Pareto(, 1) and ln(1 + Y /x0 ) exponential().

Exercise 3.4: [NLI7, Solution][?, Exercise 7] Assume Y (, c), where its density is for

y 0,

c 1

g(y) = y exp(cy). (3.1)

()

Prove the statements of the moment generating function MY and the loss size index

function I(G(y)). Hint: use the trick of the proof of Proposition 2.20 in ?.

1 I(G(u))

e(u) = Y u, E[Y 1{u1 <Y u2 } ] = Y (I(G(u2 )) I(G(u1 ))). (3.2)

1 G(u)

Exercise 3.5: [los2K, Solution] [?, Problem 2.2.1] Determine the expected value and the vari-

ance of X = IB if the claim probability equals 0.1. First, assume that B equals 5 with

probability 1. Then, let B Uniform(0,10).

Exercise 3.6: [los4K, Solution] [?, Problem 2.2.5] If X = IB, what is mX (t)?

Exercise 3.7: [los5K, Solution] [?, Problem 2.2.6] Consider the following cdf F:

0 for x < 2

x

F (x) = for 2 x < 4

4

1 for 4 x

Determine independent random variables I, X, and Y such that Z = IX + (1 I)Y has cdf

F , I Bernoulli, X is a discrete and Y a continuous random variable.

Exercise 3.8: [los6K, Solution] [?, Problem 2.2.8] Suppose that T = qX + (1 q)Y and

Z = IX + (1 I)Y with I Bernoulli(q). Compare E[T k ] with E[Z k ], k = 1, 2.

Exercise 3.9: [NLI5, Solution][?, Exercise 5] Consider the data given in Table 3.1. Estimate

the parameters for the Poisson and the negative-binomial models. Which model is preferred?

Does a 2 -goodness-of-fit test reject the null hypothesis on the 5% significance lvel of having

Poisson distributions?

t 1 2 3 4 5 6 7 8 9 10

Nt 1000 997 985 989 1056 1070 994 986 1093 1054

vt 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000

33

ACTL3162 General Insurance Techniques Exercises

Exercise 3.10: [Fit6R, Solution] The following data are the results of a sample of 250 losses:

0 - 25 5

25 - 50 37

50 - 75 28

75 - 100 31

100 - 125 23

125 - 150 9

150 - 200 22

200 - 250 17

250 - 350 15

350 - 500 17

500 - 750 13

750 - 1,000 12

1,000 - 1,500 3

1,500 - 2,500 5

2,500 - 5,000 5

5,000 - 10,000 3

10,000 - 25,000 3

25,000 2

1. Determine the maximum likelihood estimate of . (Be careful with the log-likelihood

function, since the data are grouped.)

2. Conduct a 2 goodness-of-fit test of this inverse exponential distribution model on the

data.

3. Explain the two other types of hypothesis tests, together with their similarities and differ-

ences, that can be conducted in determining the quality of the fit of your chosen model.

Exercise 3.11: [NLI9, Solution][?, Exercise 9] Assume we have i.i.d. claim sizes Y = (Y1 , ..., Yn )0

with n = 1000 which were generated by a gamma distribution, see Figure 3.1. The sample mean

and sample standard deviation are given by

bn = 0.1039 and

bn = 0.1039. (3.3)

If we fit the parameters of the gamma distribution we obtain the method of moments estimator

and the MLEs

bM M = 0.9794 and b

cM M = 9.4249, (3.4)

bM LE = 1.0013 and b

cM LE = 9.6360. (3.5)

This provides the fitted distributions displayed in Figure 3.2. The fits look perfect and the

corresponding log-likelihoods are given by

MM , b

`Y (b cM M ) = 1264.013 and `Y (b

M LE , b

cM LE ) = 1264.171. (3.6)

34

ACTL3162 General Insurance Techniques Exercises

Figure 3.1: i.i.d. claim sizes Y = (Y1 , ..., Yn )0 with n = 1000; lhs: observed data; rhs: empirical

distribution function.

Figure 3.2: Fitted gamma distribution; lhs: log-log plot; rhs: QQ plot.

M LE , b

(a) Why is `Y (b cM LE ) > `Y (b

M M , b

cM M ) and which fit should be preferred according to

AIC?

(b) The estimates of are very close to 1 and we could also use an exponential distribu-

cM LE = 9.6231

tion function. For the exponential distribution function we obtain MLE b

M LE

and `Y (b

c ) = 1264.169. Which model (gamma or exponential) should be preferred

according to the AIC and the BIC?

Exercise 3.12: [Fit7R, Solution][R ] The data in the attachment liability.txt contains data

on liability insurance claim sizes (in German former currency - Marks) for the year 1982.

2. Fit a Pareto distribution to these data using maximum likelihood estimation. Provide the

usual graphical comparisons (histogram vs. fitted parametric density function, empirical

35

ACTL3162 General Insurance Techniques Exercises

90 Chapter 3. Individual Claim Size Modeling

CDF vs. fitted parametric CDF, Q-Q plot, P-P plot).

Example 3.15 (leverage effect of claims inflation). Assume that Y0 Pareto(, )

3. Fit a Weibull distribution to these data using maximum likelihood estimation. Provide

with > 1 and choose a deductible d > . In that case we have, see (3.10),

the usual graphical comparisons.

A B

4. Fit a lognormal distribution to these data using 1

d maximum likelihood estimation. Provide

E [(Y0 d)+ ] = d.

the usual comparisons. 1

5. Which distribution would you choose? Justify your answer.

Choose inflation index i > 0 such that (1 + i) < d. From (3.7) we obtain

(d)

3.4 = (1 + i)Y

CalculatingY1 within 0 Pareto((1

layers + i), ).sizes

for claim

w)

This provides for > 1 and i > 0

Exercise 3.13: [sur12K, Solution] [?, Problem 3.8.5] Determine the cdf of Pr[Z d] and the

stop loss premium E[(Z d)+A B or combination Z of exponential distribution as

] for a mixture

d 1

in E [(Y1 d)+ ] = d

p(x) (1

= qe+ x

i) + (1 1 x , x > 0.

q)e

A B

d of Z 1z, given Z > z.

(m

Also determine the conditional = (1 + i)

distribution

d > (1 + i) E [(Y0 d)+ ] .

1

Exercise 3.14: [sur13, Solution] Show that

Observe that we obtain a strict inequality, i.e. the pure riskZ premium grows faster

than the claim sizes itself. The reason forx)dF

this(x)

faster growth is[1that

F claims

Z d d

E [(S d) + ] = E[S] d + (d = E[S] (x)] dx.Y0 d

may entitle for claims payments after 0 claims inflation adjustments,

0 i.e. not only the

claim sizes are growing under inflation but also the number of claims is growing if

tes

Exercise 3.15: [NLI10, Solution][?, Exercise 10] In Figure 3.3 we display the distribution

one does not adapt the deductible to inflation.

function of loss without reinsurance Y G and the resulting distribution function of the

loss to the insurer after applying different re-insurance covers to loss Y . Can you explicitly

Exercise

determine the 10. In Figurecovers

re-insurance 3.30 we display

from the the distribution

graphs in Figurefunction

3.3? Note of loss Y

that and

theGfunctions

belowthearedistribution

cumulative function of the

distribution loss after applying different re-insurance covers to

functions.

no

NL

Figure

Figure 3.3: 3.30: Distribution

Cumulative functions

Distribution implied

functions by re-insurance

implied contracts.

by re-insurance contracts

Y . Can you explicitly determine the re-insurance covers from the graphs in Figure

Exercise

3.30. 3.16: [NLI11, Solution][?, Exercise 11] Assume claims sizes Yi in a give line

of

business can be described by a log-normal distribution with mean E[Yi ] = 3000 and Vco(Yi ) = 4

(coefficient of variation).

Version June 29, 2015, M.V. Wthrich, ETH Zurich

36

ACTL3162 General Insurance Techniques Exercises

Up to now the insurance company was not offering contracts with deductibles. Now it wants to

offer the following three deductible versions d = 200, 500, 1000. Answer the following questions:

2. How does the expected claim size change by the introduction of deductibles?

Exercise 3.17: [sur14K, Solution] [?, Problem 3.10.4] Assume that X1 , X2 , . . . are independent

and identically distributed risks that represent the loss on a portfolio in consecutive years. We

could insure these risks with separate stop loss contracts for one year with a retention d, but

we could also consider only one contract for the whole period of n years with the retention

nd. Show that E [(X1 d)+ ] + . . . + E [(Xn d)+ ] E [(X1 + . . . + Xn nd)+ ]. If d E[Xi ],

examine how the total net stop loss premium for the one-year contracts E [(X1 d)+ ] relates

to the stop loss premium for the n-year period E [(X1 + . . . + Xn nd)+ ].

Hint [?, Rule of thumb 3.10.1]: For retentions t larger than the expectation = E[U ] = E[W ],

we have for the stop loss premiums of risks U and W :

E [(U t)+ ] V ar(U )

.

E [(W t)+ ] V ar(W )

Exercise 3.18: [sur15, Solution] [2005 Quiz 1 Question 4] An insurer has a portfolio consisting

of 1000 one year term life insurance policies that pays $100 in the event of death within one

year. The probability of death is 0.002.

The insurer has a EoL reinsurance for each policy in excess of $90.

1. For the insurer, calculate the expected total annual claims and the variance of total annual

claims without the reinsurance.

2. For the insurer, calculate the expected total annual claims and the variance of total annual

claims with the reinsurance.

3. For the reinsurer, calculate the expected total annual claims and the variance of total

annual claims.

Exercise 3.19: [sur16R, Solution][R ] This question is a follow-up from Exercise 4.18

In Exercise 4.18, we have calculated x, fS (x), FS (x), x = 0, 1, 2, . . . , 25. Now using the same

idea, prepare a table for each of the cases i = 1, 2 and 3 including:

3.5 Solutions

Solution 3.1: [los36R, Exercise] One possible code is as follows:

CMom123Gam12 <- function(cgf,param){

#initialising vectors

dcgf <- c(cgf) # for cgf, cgf, cgf, etc...

kappa <- c() # for the cumulants

37

ACTL3162 General Insurance Techniques Exercises

for(i in 1:4){

dcgf <- c(dcgf, D(dcgf[i],"t")) # i-th derivative

kappa <- c(kappa, eval(dcgf[i+1],param)) #i-th cumulant

}

# remember that 4-th cumulant is not the 4-th central moment!

gamma <- c(kappa[3]/kappa[2]^(3/2),kappa[4]/kappa[2]^2)

#returning results

c(kappa[1:3],gamma)

}

# range of rv

range <- length(pmf)-1

# descriptive statistics

m <- sum(pmf*(0:range))

v <- sum(pmf*(0:range - m)^2)

g1 <- sum(pmf*(0:range - m)^3)/v^(3/2)

g2 <- sum(pmf*(0:range - m)^4)/v^2-3

# print results

if(print==1){

results <- data.frame(mean=m,variance=v,skewness=g1,kurtosis=g2)

print(results)

}

c(m,v,g1,g2)

}

Note that it is important to return a vector with the results you may need to use later in

your code, because they are lost once the function has been processed. In other words, if

you run the function on one line, m (the mean) will not be available in the subsequent

lines. This also means (and this is the reason why it is so) that you can use m in your

main code and use the function without having a conflict between the two ms.

2. Add the following lines to the code:

source("[path to file]")

print("Descriptive statistics")

pmf_to_desc_stats(fS[,conv+1],1)

Note that [path to the file] needs to be replaced by the path to the R document where

the code of your function is. If you create several functions, you could put all of them in

a single file that is sourced at the beginning of your normal R files. This way, you can

use all your functions whenever you want. Note that before spending time creating a new

function, it may be advisable to check if this function does not already exist in R. . .

38

ACTL3162 General Insurance Techniques Exercises

E (Y j ) = mlog Y (j) . Thus,

1 2

= exp +

2

and

E Y2

= E [exp (2 log Y )] = mlog Y (2)

= exp 2 + 2 2

so that

V ar (Y ) = exp 2 + 2 exp 2 1 .

If Y Pareto(, x0 ) then xY0

Pareto(, 1) and log 1 + xY0 Exponential ().

Proof

Y

Pr y = Pr [Y yx0 ]

x0

x0

=1

x0 + yx0

1

=1 which is the CDF of a Pareto (, 1) random variable

(1 + y)

and

Y

Pr ln 1 + y = Pr [Y (ey 1)x0 ]

x0

x0

=1

x0 + x0 (ey 1)

= 1 ey which is the CDF of an Exp() random variable

So

E [Y ] = E [Y + x0 ] x0

Y

= x0 E + 1 x0

x0

Y

= x0 E exp 1 ln +1 x0

x0

= x0 MZ (1) x0 where Z Exponential()

= x0 x0

1

x0

=

1

Now

Y

E exp 2 ln +1 = MZ (2) where Z Exponential()

x0

=

2

39

ACTL3162 General Insurance Techniques Exercises

But

" 2 #

Y Y + x0

E exp 2 ln +1 =E

x0 x0

1

= 2 E Y 2 + 2Y x0 + x20

x0

So

2 x0

2 2

E Y = x0 x0 2x0

2 1

2 ( 1) ( 1)( 2) 2( 2)

= x0

( 1)( 2)

2

x0

= (2( 1) 2( 2))

( 1)( 2)

2x20

=

( 1)( 2)

Then

2

2x20

x0

V ar [Y ] =

( 1)( 2) 1

2

2x0 ( 1) x20 ( 2)

=

( 1)2 ( 2) ( 1)2 ( 2)

x20 (2 2 + 2)

=

( 1)2 ( 2)

x20

=

( 1)2 ( 2)

Z

c 1

MY (r) = exp(ry) y exp(cy)dy

0 ()

Z

(c r) 1

c

= y exp[(c r)y]dy

cr 0 ()

c

= . (3.7)

cr

Next we find the loss size index function for level y, I(G(y))

I(G(y)) = E[Y 1{Y y} ]/Y

c y c 1

Z

= x x exp(cx)dx

0 ()

Z y

c+1

= x+11 exp(cx)dx

0 ( + 1)

Z cy

1

= z +11 exp(z)dz

( + 1) 0

= G( + 1, cy). (3.8)

40

ACTL3162 General Insurance Techniques Exercises

Z

yg(y)

= dy u

u 1 G(u)

Z Z u

yg(y) yg(y)

= dy dy u

0 1 G(u) 0 1 G(u)

E[Y 1{Y u} ]

= Y u

1 G(u)

I(G(u))

= Y 1 u (3.9)

1 G(u)

= E[Y 1{Y u2 } ] E[Y 1{Y u1 } ]

= Y (I(G(u2 )) I(G(u1 ))). (3.10)

Solution 3.5: [los2K, Exercise] First note that with B = 5 w.p. 1, we have = E [B |I = 1] =

5 and 2 = Var[B|I = 1] = 0. We will use the decomposition of variance result that Var[X] =

Var[E[B|I]] + E[Var[B|I]]. Below, we also have that q = 0.1, which is the probability of a claim

occurring. It is then clear that

E (X) = q = 0.5

and

V ar (X) = 2 q (1 q) + 2 q = 9/4.

100

Now, if B U (0, 10), then = 5 and 2 = 12

. Thus,

E (X) = q = 0.5

and

V ar (X) = 2 q (1 q) + 2 q = 37/12.

mX (t) = E E eXt |I

= (1 q) e0 + qmX|I=1 (t)

= 1 q + qmB (t) .

Solution 3.7: [los5K, Exercise] F (x) has a jump of size 1/2 in 2 and is uniform on (2, 4), so

F is the following mixture of cdfs:

1 1

F (x) = G (x) + H (x)

2 2

with dG (2) = 1 and H Uniform(2, 4). The mixed r.v. I X + (1 I) Y has cdf F for I

Bernoulli(1/2), X 2 and Y Uniform(2, 4), independent.

41

ACTL3162 General Insurance Techniques Exercises

E (T ) = qE (X) + (1 q) E (Y )

and

E T2 = E q 2 X 2 + 2q (1 q) XY + (1 q)2 Y 2

= q 2 E X 2 + 2q (1 q) E (X) E (Y ) + (1 q)2 E Y 2 .

E (Z) = E [IX + (1 I) Y ]

= E (I) E (X) + (1 E (I)) E (Y )

= qE (X) + (1 q) E (Y ) = E (T )

and

E Z2 = E I 2 X 2 + 2I (1 I) XY + (1 I)2 Y 2

= E I 2 E X 2 + 0 + E (1 I)2 E Y 2

= qE X 2 + (1 q) E Y 2 .

Solution 3.9: [NLI5, Exercise] Obtaining MLE for Poisson is easy and the formula is

T

1 X

M LE

P OI = PT Nt . (3.11)

t=1 vt t=1

T

X (Nt M LE vt )2

P OI

2P OI = M LE

= 14.83803. (3.12)

t=1

P OI vt

We compare this value to the 95%-quantile of the Chi-square distribution with 9 degrees of

freedom 16.91898. Since the value of test statistic is smaller than the critical value, we cannot

reject the null hypothesis on the 5% significance level.

For the negative-binomial model, we do not have explicit formula for the MLE and we use

the function fitdistr(...) from the MASS package in R. Note that we can do this since vt

have uniform values of 10000, so we can find the maximum likelihood estimator for the size

parameter and divide by 100000. Theen test statistic for the negative binomial model is similar

to the Poisson one,

T

X (Nt M LE

N B vt )

2

2N B = M LE

= 1955.112. (3.13)

t=1

N B vt

This is way larger than the 95%-quantile of the Chi-square distribution with 9 degrees of

freedom 16.91898. So it is clear that the Poisson model is preferred.

R-code:

42

ACTL3162 General Insurance Techniques Exercises

# read the table first

N.vector<-c(1000,997,985,989,1056,1070,994,986,1093,1054)

v.vector<-rep(10000,10)

# Poisson MLE estimator of Lambda

lambda.Poi<-sum(N.vector)/sum(v.vector)

# chi square test statistics

sum((N.vector-lambda.Poi*v.vector)^2/(lambda.Poi*v.vector))

# we then need the 95%-quantile of the chi-sq with 9 degrees of freedom

qchisq(0.95,9)

library(MASS)

# use fitdistr(...) function to find the MLE

NB.fit<-fitdistr(N.vector,"negative binomial")

lambda.NB<-NB.fit$estimate[1]/10000

# chi square test statistics

sum((N.vector-lambda.NB*v.vector)^2/(lambda.NB*v.vector))

Solution 3.10: [Fit6R, Exercise] Denote the range of losses by [ai , bi ) for i = 1, 2, ...18 since

there are 18 given intervals of losses. Note that for a1 = 0 and b18 = .

18

Y

L (; ai , bi , ni ) = [F (bi ) F (ai )]ni

i=1

where ni refers to the number of claims observed in the i-th range of loss. The log-

likelihood is given by

18

X

` (; ai , bi , ni ) = ni log [exp (/bi ) exp (/ai )] .

i=1

` (; ai , bi , ni )

18

X ni b1i exp (/bi ) + 1

ai

exp (/ai )

= =0

i=1

exp (/bi ) exp (/ai )

which can only be numerically computed. So we will use Excel to find the MLE of . In

the sheet Parameter Estimate of Fit6R.xls(x), the derivatives are computed and use

goal seek to find the MLE such that the derivative is zero. The parameter estimate for

is 93.18568.

2. In the sheet Chi-sq Test of Fit6R.xls(x), we use Excel to compute the Chi-square test

statistic 16.5232 and the associated p-value 0.34825.

It is clear that the result of the chi-square test above does support the Inverse Exponential

model for the given data. The two other tests that can be conducted are the Kolmogorov-

Smirnoff (K-S) and the Anderson-Darling (A-D) tests. Both test, similar to the chi-square

43

ACTL3162 General Insurance Techniques Exercises

test, whether the data comes from the assumed population. K-S and A-D tests are quite

similar - both look at the difference between the empirical and model dfs: K-S in absolute

value, A-D in squared difference. But A-D is weighted average, with more emphasis on

good fit in the tails than in the middle; K-S puts no such emphasis. For K-S and A-D

tests, no adjustments are made to account for increase in the number of parameters. The

result is that more complex models often will fare better on these tests. On the other

hand, the 2 test adjusts the d.f. for increases in the number of parameters. All 3 tests

are sensitive to sample size:

large sample size therefore increases probability of rejecting all models

cM LE ) jointly maximise the likelihood func-

tion, which also maximise the log-likelihood function. This is the reason why we have

M LE , b

`Y (b cM LE ) > `Y (b

MM , b

cM M ). For both methods, two estimated parameters are

involved, so the MLE approach produces the smallest AIC value and is preferred.

(b) We calculate the AIC and the BIC for gamma and exponential fits,

BICexp = 2521.43 and BICgamma = 2514.53. (3.15)

According to both the AIC and the BIC criteria, exponential distribution is preferred

(smallest AIC/BIC).

data<-read.table("liability.txt",header=T)

attach(data)

source("DataSummStats.R")

DataSummStats(liability)

Value

Number 6.800000e+02

Mean 7.416837e+04

5th Q 3.000000e+04

25th Q 3.500000e+04

Median 5.000000e+04

75th Q 7.500000e+04

95th Q 2.000000e+05

Variance 8.689669e+09

StdDev 9.321839e+04

Minimum 3.000000e+04

Maximum 1.200000e+06

Skewness 6.440000e+00

Kurtosis 5.720000e+01

44

ACTL3162 General Insurance Techniques Exercises

f (x) = , x1 , ..., xn > .

x+1

Using the maximum likelihood estimation, we can obtain

n

mle = Pn .

i=1 (log(xi ) log())

Substituting the data yields, mle = 30000 and mle = 2.012135. The graphical compar-

isons are shown below.

source("qpareto.R")

source("dpareto.R")

source("ppareto.R")

#MLE

lambda.hat<-min(liability)

alpha.hat<-860/sum(log(liability)-log(lambda.hat))

par.hat.p<-c(alpha.hat,lambda.hat)

par.hat.p

par(mfrow=c(2,2))

#histogram

hist(liability,breaks=100,prob=T,xlab="claims",main="Histogram of Claims")

xgrid<-seq(min(liability),max(liability),length=860)

lines(xgrid,dpareto(xgrid,par.hat.p[1],par.hat.p[2]),col=2)

legend(300000,0.00002,legend=c("Pareto Model"),lty=1,col=2)

empirical<-ecdf(liability)

plot(xgrid,empirical(xgrid),type="l",xlab="claims",ylab="cdf",main="Empirical CDF")

lines(xgrid,ppareto(xgrid,par.hat.p[1],par.hat.p[2]),col=2)

legend(300000,0.6,legend=c("Empirical cdf","Estimated cdf"),lty=1,col=1:2)

#qqplot

plot(qpareto(empirical(liability),par.hat.p[1],par.hat.p[2]),liability,

xlab="theoretical quantiles",ylab="sample quantiles",main="Q-Q plot",cex=0.45)

abline(0,1,col=2)

#ppplot

plot(ppareto(liability,par.hat.p[1],par.hat.p[2]),empirical(liability),

xlab="theoretical probability",ylab="sample probability",main="P-P plot",cex=0.45)

abline(0,1,col=2)

45

ACTL3162 General Insurance Techniques Exercises

k

F (x) = 1 e(x/) , x 0.

mle = 79791.605439

kmle = 1.184008.

loglike<-function(x,par){

-sum(log(dweibull(x,par[1],par[2])))

}

fit.wei<-constrOptim(init.est,loglike,NULL,ui=c(1,1),ci=c(0,0),x=liability)

par.hat.w<-fit.wei$par

par.hat.w

par(mfrow=c(2,2))

#histogram

hist(liability,breaks=100,prob=T,xlab="claims",main="Histogram of Claims")

xgrid<-seq(min(liability),max(liability),length=860)

46

ACTL3162 General Insurance Techniques Exercises

lines(xgrid,dweibull(xgrid,par.hat.w[1],par.hat.w[2]),col=2)

legend(300000,0.00002,legend=c("Weibull Model"),lty=1,col=2)

empirical<-ecdf(liability)

plot(xgrid,empirical(xgrid),type="l",xlab="claims",ylab="cdf",main="Empirical CDF")

lines(xgrid,pweibull(xgrid,par.hat.w[1],par.hat.w[2]),col=2)

legend(300000,0.6,legend=c("Empirical cdf","Estimated cdf"),lty=1,col=1:2)

#qqplot

plot(qweibull(empirical(liability),par.hat.w[1],par.hat.w[2]),liability,

xlab="theoretical quantiles",ylab="sample quantiles",main="Q-Q plot",cex=0.45)

abline(0,1,col=2)

#ppplot

plot(pweibull(liability,par.hat.w[1],par.hat.w[2]),empirical(liability),

xlab="theoretical probability",ylab="sample probability",main="P-P plot",cex=0.45)

abline(0,1,col=2)

4. For the log-normal case, there are two parameters need to be estimated ( and 2 ). Use

47

ACTL3162 General Insurance Techniques Exercises

mle = 10.9373482

mle = 0.6262126

loglike<-function(x,par){

-sum(log(dlnorm(x,par[1],par[2])))

}

fit.ln<-constrOptim(init.est,loglike,NULL,ui=c(1,1),ci=c(0,0),x=liability)

par.hat.l<-fit.ln$par

par.hat.l

par(mfrow=c(2,2))

#histogram

hist(liability,breaks=100,prob=T,xlab="claims",main="Histogram of Claims")

xgrid<-seq(min(liability),max(liability),length=860)

lines(xgrid,dlnorm(xgrid,par.hat.l[1],par.hat.l[2]),col=2)

legend(300000,0.00002,legend=c("Lognormal Model"),lty=1,col=2)

empirical<-ecdf(liability)

plot(xgrid,empirical(xgrid),type="l",xlab="claims",ylab="cdf",main="Empirical CDF")

lines(xgrid,plnorm(xgrid,par.hat.l[1],par.hat.l[2]),col=2)

legend(300000,0.6,legend=c("Empirical cdf","Estimated cdf"),lty=1,col=1:2)

#qqplot

plot(qlnorm(empirical(liability),par.hat.l[1],par.hat.l[2]),liability,

xlab="theoretical quantiles",ylab="sample quantiles",main="Q-Q plot",cex=0.45)

abline(0,1,col=2)

#ppplot

plot(plnorm(liability,par.hat.l[1],par.hat.l[2]),emprical(liability),

xlab="theoretical probability",ylab="sample probability",main="P-P plot",cex=0.45)

abline(0,1,col=2)

48

ACTL3162 General Insurance Techniques Exercises

5. #loglikelihood

sum(log(dpareto(liability,par.hat.p[1],par.hat.p[2])))

sum(log(dweibull(liability,par.hat.w[1],par.hat.w[2])))

sum(log(dlnorm(liability,par.hat.l[1],par.hat.l[2])))

#kolmogorov-smirnoff

max(abs(ecdf(liability)(liability)-ppareto(liability,par.hat.p[1],par.hat.p[2])))

max(abs(ecdf(liability)(liability)-pweibull(liability,par.hat.w[1],par.hat.w[2])))

max(abs(ecdf(liability)(liability)-plnorm(liability,par.hat.l[1],par.hat.l[2])))

#A-D statistics

f.p<-function(x){

860*dpareto(x,par.hat.p[1],par.hat.p[2])*(ecdf(liability)(x)

-ppareto(x,par.hat.p[1],par.hat.p[2]))^2/

(ppareto(x,par.hat.p[1],par.hat.p[2])*(1-ppareto(x,par.hat.p[1],par.hat.p[2])))

}

f.w<-function(x){

860*dweibull(x,par.hat.w[1],par.hat.w[2])*(ecdf(liability)(x)

-pweibull(x,par.hat.w[1],par.hat.w[2]))^2/

(pweibull(x,par.hat.w[1],par.hat.w[2])*(1-pweibull(x,par.hat.w[1],par.hat.w[2])))

}

f.l<-function(x){

860*dlnorm(x,par.hat.l[1],par.hat.l[2])*(ecdf(liability)(x)

-plnorm(x,par.hat.l[1],par.hat.l[2]))^2/

49

ACTL3162 General Insurance Techniques Exercises

(plnorm(x,par.hat.l[1],par.hat.l[2])*(1-plnorm(x,par.hat.l[1],par.hat.l[2])))

}

sum(f.p(seq(30001,1200000)))

sum(f.w(seq(30001,1200000)))

sum(f.l(seq(30001,1200000)))

#Chi-square statistics

n=100

interval<-seq(30000,1200000,length=n)

e.p<-c()

for(i in 1:n-1){

e.p[i]<-860*(ppareto(interval[i+1],par.hat.p[1],par.hat.p[2])

-ppareto(interval[i],par.hat.p[1],par.hat.p[2]))

}

e.w<-c()

interval<-seq(30000,1200000,length=n)

for(i in 1:n-1){

e.w[i]<-860*(pweibull(interval[i+1],par.hat.w[1],par.hat.w[2])

-pweibull(interval[i],par.hat.w[1],par.hat.w[2]))

}

e.l<-c()

for(i in 1:n-1){

e.l[i]<-860*(plnorm(interval[i+1],par.hat.l[1],par.hat.l[2])

-plnorm(interval[i],par.hat.l[1],par.hat.l[2]))

}

o<-c()

for(i in 1:n-1){

o[i]<-860*(ecdf(liability)(interval[i+1])-ecdf(liability)(interval[i]))

}

sum((e.p-o)^2/e.p)

sum((e.w-o)^2/e.w)

sum((e.l-o)^2/e.l)

We will evaluate the models based on the following table.

Model Loglikelihood K-S A-D 2

Pareto -7822.041 0.1820444 205.7854 95384.95

Weibull -8284.174 0.1876902 30.54793 37484626594

Lognormal -8083.975 0.1968117 21.46748 6058129

Using the table above, we will choose the Pareto distribution as it has the highest log-

likelihood, lowers K-S and 2 test statistics. But its high value of A-D statistic is quite

baffling and therefore not considered.

We get the same mixtures or combinations of the cdfs and slps with exponential distributions:

P (Z > d) = qed + (1 q) ed

E (Z d)+ = qed / + (1 q) ed /.

P (Z z > y |Z > z ) = q (z) ey + (1 q (z)) ey

with q (z) the following function:

q (z) = qez / qez + (1 q) ez .

50

ACTL3162 General Insurance Techniques Exercises

Z

E[(S d)+ ] = (x d)f (x)dx

d

Z Z

= xf (x)dx d f (x)dx

d d

Z Z d Z Z d

= xf (x)dx xf (x)dx d f (x)dx + d f (x)dx

0 0 0 0

Z d

= E[S] d + (d x)dF (x)

0

Z d

d

= E[S] d + [(d x)F (x)]x=0 F (x)d(d x)

0

Z d

= E[S] d + F (x)d(x)

0

Z d

= E[S] [1 F (x)]dx

0

Note that G(y) gives the cumulative distribution of the losses without reinsurance.

First for re-insurance type 1, there is a mass at zero (blue line) and this probability mass

corresponds to GY (1). This means that re-insurance type 1 is a deductible at level 1, (Y 1)+ .

For re-insurance type 2, the probability mass beyond y = 2 is zero and the distribution function

remains the same as G(y) below y = 2. Both imply that the re-insurance type 2 is a non-

proportional reinsurance contract with policy limit 2, Y 2.

For re-insurance type 3, the distribution function is the same as the original below y = 1, so

there is no re-insurance contract for this section. There is a mass at y = 2 (discontinuity in red

line) and the resulting distribution function corresponds to GY (2). This means that for claims

of size larger than 1, the re-insurance type 3 provides a deductible of level 1.

2. With the introduction of deductibles, the expected claim size will decrease.

3. Recall that E[(Y d)+ ] = P[Y > d]e(d), where d > 0 is the deductible level and e(d) is

the mean excess above d. For a log-normal distributed claim,

log d ( + 2 )

log d

E[(Y d)+ ] = Y 1 d 1 . (3.16)

Substituting d = 200, 500, 1000 yields E[(Y 200)+ ] = 2822.893, E[(Y 500)+ ] = 2620.95

and E[(Y 1000)+ ] = 2372.275.

R-codes:

51

ACTL3162 General Insurance Techniques Exercises

mean<-3000

vco<-4

sigma2<-log(vco^2+1)

sigma<-sigma2^(1/2)

mu<-log(mean)-sigma2/2

d<-200

mean*(1-pnorm((log(d)-mu-sigma2)/sigma))-d*(1-pnorm((log(d)-mu)/sigma))

d<-500

mean*(1-pnorm((log(d)-mu-sigma2)/sigma))-d*(1-pnorm((log(d)-mu)/sigma))

d<-1000

mean*(1-pnorm((log(d)-mu-sigma2)/sigma))-d*(1-pnorm((log(d)-mu)/sigma))

X X X X

xi nd = (xi d) (xi d)+ = (xi d)+ .

+ + +

X h i

E xi nd = nE X d + .

+

Also,

1

V ar X = V ar (X1 ) .

n

Using the hint, for d ,

h i V ar X

E X d + /E (X1 d)+ .

V ar (X1 )

This leads to P

E ( xi nd)+

1.

E (X1 d)+

Hence, the slp for a one-year contract with retention d is about as large as the one for an n-year

period with retention nd.

Solution 3.18: [sur15, Exercise] Let S be the total annual claims paid by the insurer in the

absence of the stop-loss reinsurance, S be the total annual claims paid by the insurer in the

presence of the stop-loss reinsurance, and SR be the total annual claims paid by the reinsurance.

We have:

52

ACTL3162 General Insurance Techniques Exercises

Solution 3.19: [sur16R, Exercise] For the solution code, refer to solution of [los40R, Exercise]

1. Below are tables for d, E[(S d)+ ], E[((S d)+ )2 ], V ar[(S d)+ ], d = 0, 1, 2, . . . , 25.

[1] "Distribution of stop-loss premiums if N is Poisson"

print(Poisson[,4:7],print.gap=3)

d E[(S-d)+] E[(S-d)+^2] Var[(S-d)+]

0 18.0000000000 454.000000000 130.000000000

1 17.0407622040 418.959237796 128.571661303

2 16.0978292895 385.820646302 126.680538467

3 15.1663097922 354.556507221 124.539554507

4 14.2547909497 325.135406479 121.936341460

5 13.3594574196 297.521158110 119.046055564

6 12.4886716496 271.673029040 115.706109469

7 11.6397107390 247.544646652 112.061780564

8 10.8208210870 225.084114826 107.993945829

9 10.0303982617 204.232895477 103.624006188

10 9.2766869917 184.925810224 98.868888682

11 8.5673901552 167.081733077 93.681559005

12 7.8896753530 150.624667569 88.377690393

13 7.2449475938 135.490044622 83.000778985

14 6.6340789981 121.611018030 77.600013876

15 6.0574949924 108.919444039 72.226198457

16 5.5162157850 97.345733262 66.917096675

17 5.0096373915 86.819880085 61.723413291

18 4.5387041817 77.271538512 56.671702863

19 4.1017018776 68.631132453 51.807174160

20 3.6992547622 60.830175813 47.145690017

21 3.3292385667 53.801682484 42.717853050

22 2.9879389459 47.484504971 38.556725827

23 2.6745280175 41.822038008 34.668937892

24 2.3877410250 36.759768966 31.058461763

25 2.1264373905 32.245590550 27.723854574

[1] "Distribution of stop-loss premiums if N is Negative Binomial"

print(NegBin[,4:7],print.gap=3)

d E[(S-d)+] E[(S-d)+^2] Var[(S-d)+]

0 18.00000000 535.0000000 211.0000000

1 17.09525987 499.9047401 207.6568301

53

ACTL3162 General Insurance Techniques Exercises

3 15.34164188 435.0444612 199.6784858

4 14.49603081 405.2067885 195.0718792

5 13.66814123 377.0426165 190.2245318

6 12.86859214 350.5058831 184.9052195

7 12.09161076 325.5456802 179.3386293

8 11.34760558 302.1064638 173.3383114

9 10.63176435 280.1270939 167.0926806

10 9.95437163 259.5409579 160.4514433

11 9.32218787 240.2643984 153.3612117

12 8.71665314 222.2255574 146.2455154

13 8.13869304 205.3702112 139.1318868

14 7.58859823 189.6429200 132.0560968

15 7.06644148 174.9878802 125.0532850

16 6.57275805 161.3486807 118.1475323

17 6.10673999 148.6691827 111.3769094

18 5.66900677 136.8934359 104.7557981

19 5.25781821 125.9666109 98.3219586

20 4.87370111 115.8350916 92.0821292

21 4.51463981 106.4467507 86.0647781

22 4.17744116 97.7546697 80.3036551

23 3.86159068 89.7156379 74.8037554

24 3.56626265 82.2877846 69.5695553

25 3.29074389 75.4307780 64.6017827

[1] "Distribution of stop-loss premiums if N is Binomial"

print(Bin[,4:7],print.gap=3)

d E[(S-d)+] E[(S-d)+^2] Var[(S-d)+]

0 1.800000e+01 4.135000e+02 8.950000e+01

1 1.701680e+01 3.784832e+02 8.891185e+01

2 1.604479e+01 3.454216e+02 8.798634e+01

3 1.508165e+01 3.142952e+02 8.683907e+01

4 1.413351e+01 2.850800e+02 8.532380e+01

5 1.319920e+01 2.577473e+02 8.352842e+01

6 1.228477e+01 2.322633e+02 8.134766e+01

7 1.139022e+01 2.085883e+02 7.885132e+01

8 1.052165e+01 1.866765e+02 7.597125e+01

9 9.680273e+00 1.664745e+02 7.276686e+01

10 8.872400e+00 1.479219e+02 6.920238e+01

11 8.106065e+00 1.309434e+02 6.523512e+01

12 7.374416e+00 1.154629e+02 6.108090e+01

13 6.679437e+00 1.014091e+02 5.679419e+01

14 6.022823e+00 8.870681e+01 5.243241e+01

15 5.405522e+00 7.727846e+01 4.805880e+01

16 4.829187e+00 6.704376e+01 4.372271e+01

17 4.293645e+00 5.792092e+01 3.948554e+01

18 3.800209e+00 4.982707e+01 3.538548e+01

19 3.347468e+00 4.267939e+01 3.147385e+01

20 2.936074e+00 3.639585e+01 2.777532e+01

21 2.564032e+00 3.089575e+01 2.432149e+01

22 2.227323e+00 2.610439e+01 2.114343e+01

23 1.924682e+00 2.195239e+01 1.824799e+01

24 1.654269e+00 1.837344e+01 1.563683e+01

25 1.414363e+00 1.530480e+01 1.330438e+01

assumption. If the portfolio has a bigger variance(more risk), then the stop-loss premium

will be higher. Hence we will expect the most stop-loss premium for the most variable

54

ACTL3162 General Insurance Techniques Exercises

portfolio, i.e. the negative binomial case. We will expect the least stop-loss premium for

the least variable portfolio, i.e. the binomial case. Illustration by the diagram below.

Stoploss Premium

15

E[(Sd)+]

10

Poi

NegBin

Bin

5

0

0 20 40 60 80

55

Module 4

Distributions

4.1 Approximations

Exercise 4.1: [los29, Solution] [2004 Final Exam Question] For a specific type of insurance

risk, the claim amount X is being modeled as the product of two random variables I and B as

X = I B,

where I is the claim indicator random variable and B is the random variable representing the

amount of the claim, conditional on the event that a claim occurs.

Now, assume that the probability that a claim occurs is P (I = 1) = q, or otherwise, the

probability that no claim occurs is P (I = 0) = 1 q. Furthermore, denote by and 2 to be

the mean and the variance of B, respectively.

B =bZ

where b is a fixed constant and Z is a Poisson random variable with parameter , so that

B takes possible values of 0, b, 2b, 3b, and so on. Determine expressions for the mean,

variance, and probability distribution of X in terms of the parameters , q, and b.

Type Random Variable

A XA b = 1, = 1.5, q = 0.1

B XB b = 2, = 1.0, q = 0.2

Given that the two types of risk are independent, use the convolution formula to calculate

the probability that XA + XB is greater than 3. Do not use any approximations.

56

ACTL3162 General Insurance Techniques Exercises

4. Consider a portfolio of 100 independent Type A risks (above). Use the Normal (Central

Limit Theorem) approximation to calculate the probability that the average claim amount

per risk is greater than 0.5.

Exercise 4.2: [NLI12, Solution][?, Exercise 12] Assume that S has a compound Poisson dis-

tribution with expected number of claims v > 0 and claim size distribution G having finite

third moment.

1. Prove that the fit of moment approximation for a translated gamma distribution for X

provides the following system of equations

E[Y13 ]

vE[Y1 ] = k + /c, vE[Y12 ] = /c2 and = 2 1/2 . (4.1)

(v)1/2 E[Y12 ]3/2

2. Solve this system of equations for k R, > 0 and c > 0 (assume that G(0) = 0).

Exercise 4.3: [los31K, Solution] [?, Problem 3.6.1] Assume S is compound Poisson distributed

with parameter = 12 and uniform(0, 1) distributed claims. Approximate Pr[S < 10] with the

CLT approximation, the translated gamma approximation and the NP approximation.

Exercise 4.4: [los32K, Solution] [?, Problem 2.5.9] An insurers portfolio contains 2000 one-

year life insurance policies. Half of them are characterised by a payment b1 = 1 and a probability

of dying within 1 year of q1 = 1%. For the other half, we have b2 = 2 and q2 = 5%. Use the CLT

to determine the minimum safety loading, as a percentage, to be added to the net premium

to ensure that the probability that the total payment exceeds the total premium income is at

most 5%.

Exercise 4.5: [los33K, Solution] [?, Problem 2.5.13] A portfolio consists of two types of con-

tracts. For type k, k = 1, 2, the claim probability is qk and the number of policies is nk . If

there is a claim, then its size is x with probability pk (x):

Type 1 1000 0.01 0.5 0 0.5

Type 2 2000 0.02 0.5 0.5 0

Assume that the contracts are independent. Let Sk denote the total claim amount of the

contracts of type k and let S = S1 + S2 . Calculate the expected value and the variance of a

contract of type k, k = 1, 2. Then, calculate the expected value and the variance of S. Use the

CLT to determine the minimum capital that covers all claims with probability 95%.

Exercise 4.6: [los34, Solution] We want to approximate the individual model Se by a collective

model S. Show that j = ln(1 qj ) yields both a larger expectation and a larger variance

than the ones of S.

e For both cases, compare Pr[Ii = j] and Pr[Ni = j], j = 0, 1, 2, . . . in both

models as well as the cdfs of Ii and Ni .

Exercise 4.7: [los35K, Solution] [?, Problem 3.7.2] Consider a portfolio of 100 one-year life

insurance policies. 25 policies have insured amounts 1 and probability dying within this year

0.01, 25 policies have insured amounts 1 and probability dying within this year 0.02, 25 policies

have insured amounts 2 and probability dying within this year 0.01 and 25 policies have insured

amounts 2 and probability dying within this year 0.02.

57

ACTL3162 General Insurance Techniques Exercises

Determine the expectation and the variance of the total claims S. e Choose an appropriate

compound Poisson distribution S to approximate S and compare the expectations and the

e

variances. Determine for both S and Se the parameters of a suitable approximating translated

gamma distribution.

Exercise 4.8: [los37R, Solution][R ] Develop a function that will return and print (if the user

wishes so) the following approximations of Pr[S s] in function of the first three central

moments, as well as 1 and 2 :

Normal Power 2

translated gamma

" #

S E[S] 1 2 2

Pr p z (z) (3) (z) + (4) (z) + 1 (6) (z),

V ar(S) 6 24 72

dk

(k) (x) = (x).

dxk

Note

1 1 2

(1) (x) = (x) = e 2 x

2

(2) (x) = x(x)

(3) (x) = (x2 1)(x)

etc. . .

exercise) with s = 0.6 should yield:

NP1 NP2 EW1 EW2 EW3 TG

Pr[X<=0.6] 0.6113513 0.7037253 0.7360268 0.8349541 0.7386938 0.728984

This question is a follow-up Exercise 4.18

1. For each of the three cases i = 1, 2, 3, give the true value of Pr(Si > 50) as well as its

CLT, translated gamma and normal power approximations.

58

ACTL3162 General Insurance Techniques Exercises

Exercise 4.10: [los24, Solution] Determine a and b for the (a, b) family (Poisson, negative

binomial and binomial), also called Panjer distributions (see MW Definition 4.6).

1, 2, 3, 4. Determine fS (s) for s = 0, 1, 2, 3, 4. using Panjers recursion algorithm.

Exercise 4.12: [los26, Solution] Check the results of Exercise 2.5 using Panjers recursion

algorithm.

Exercise 4.13: [los27, Solution] The individual claim amount distribution has the following

distribution:

x P (X = x)

1 0.2

2 0.2

3 0.2

4 0.4

Determine fS (s) for s = 0, 1, 2, 3, 4:

Exercise 4.14: [los28, Solution] Let p(x) be Poisson(1). Verify that pn (x) using de Prils

algorithm will yield the exact probabilities of a Poisson(n) random variable for x = 0, 1, 2, 3.

Exercise 4.15: [los30R, Solution][R ] Develop a function for Panjers recursion that returns

fS (x) and FS (x), and this in function of a, b, fX (x) and fS (0). In addition, the function should

allow for the choice (type below) between computing

whatever number of values that is necessary in order to have all FS (x) < 1 and the

first > 1 . ( vartype below)

Finally, the function should also allow its user to print fS (x) and FS (x) for all the recursions

that were performed (binary print below). The function could then begin in the following

way:

[code omitted ^]

Exercise 4.16: [los38R, Solution][R ] Develop a function that will discretise a continuous dis-

tribution in m steps of length h, in function of its pdf or cdf. One possible beginning is

discretisation <-function(densityorcdf,type,h,m){

59

ACTL3162 General Insurance Techniques Exercises

where densityorcdf is a function and where type is binary and defines if densityorcdf if

a cdf (1) or a density (0).

The following codediscretisation of a gamma( = 2, = 4)

pmf<-discretisation(function(x){16*x*exp(-4*x)},0,0.002,1500)

plot(pmf)

0.0030

0.0025

0.0020

0.0015

pmf

0.0010

0.0005

0.0000

Index

Exercise 4.17: [los39R, Solution][R ] Let S compound Poisson( = 20, p(x) = 0.2e0.2x ).

We are interested in determining Pr[S 150]. Use the functions developed in the previous

exercises to compute all 6 approximations mentioned in Exercise 4.8, as well as the equivalent

probability using Panjers recursion after having discretised p(x) with h = 0.01 and m =

5000. Calculate the deviation between the approximations and the probability calculated with

Panjers recursion.

Here are the outputs you should get:

Approximations for Pr[X<=150]:

NP1 NP2 EW1 EW2 EW3 TG Panjer

Pr[X<=150] 0.9430769 0.93132 0.9295227 0.9306522 0.93277 0.9325275 0.9323526

NP1 NP2 EW1 EW2 EW3 TG Panjer

Pr[X<=150] 0.01072422 -0.001032627 -0.002829964 -0.001700449 0.0004173918 0.0001748720 0

Let

0.2 if x = 0,

0.1 if x {1, 3, 5, 7, 9, 10},

Pr(Xj = x) =

0.05 if x {2, 4, 6, 8}, and

0 otherwise,

and

Si = X1 + X2 + . . . + XNi , i = 1, 2, 3

60

ACTL3162 General Insurance Techniques Exercises

with

N1 Poisson(1 = 4),

N2 Neg Bin(r2 = 4, p2 = 0.5), and

N3 Bin(n3 = 8, p3 = 0.5),

1. Using Panjers recursion formula, prepare a table for each of the cases i = 1, 2 and 3

including:

x, fS (x), FS (x), x = 0, 1, 2, . . . , 25

E[()k ], k = 1, 2, 3

E[( E[])k ], k = 2, 3

1 ()

For the moments of Si , use the distribution of Si you computed with Panjers recursion

formula (after an appropriate number of recursions).

3. Interpret your results in part 1 and 2 (compare the three assumptions, i = 1, 2 or 3).

4.3 Solutions

Solution 4.1: [los29, Exercise] This was a Year 2004 final exam question.

E (X) = E [E (X |I )] = E (X |I = 1) P (I = 1) = qE (B) = q

V ar (X) = V ar [E (X |I )] + E [V ar (X |I )]

= [E (B)]2 V ar (I) + qV ar (B)

= q (1 q) [E (B)]2 + qV ar (B)

V ar (X |I = 0) = 0. Thus, we have

V ar (X) = q (1 q) 2 + q 2 .

E (X) = qb

and

V ar (X) = q (1 q) b2 2 + qb2 .

61

ACTL3162 General Insurance Techniques Exercises

mX (t) = E[etx ]

= E[etIB ]

= E[etIB |I = 1]P (I = 1) + E[etIB |I = 0]P (I = 0)

= E[etIB ]q + 1 (1 q)

bt 1)

= qe(e + (1 q)

X k ekbt

= qe + (1 q)

k=0

k!

X k kbt

= (1 q) + qe + qe e .

k=1

k!

The distribution of X can be identified by using the 1-1 correspondence between distri-

bution and mgf, i.e. by matching

X

mX (t) = E[etX ] = ety P (t = y)

y=0

with the expression above. Thus, by carefully selecting coefficients of eXt , we find

P (X = 0) = (1 q) + qe

and

x

P (X = bx) = qe , for x = 1, 2, 3, ...

x!

3. Notice that XA takes possible values 0, 1, 2, ... while XB takes possible values 0, 2, 4, ...

Now, doing the convolution for XA + XB , we have

(1) (2) (3) (4) (6)

x P (XA = x) P (XB = x) P (XA + XB = x) P (XA + XB x)

= (2)*(3)

0 0.922313 0.873576 0.805710 0.805710

1 0.033470 0 0.029238 0.834949

2 0.025102 0.073576 0.089789 0.924737

3 0.012551 0 0.013427 0.938164

i 1.5 1.0

x x/2

qe1.5 (1.5)

x!

qe1 (1)

(x/2)!

0.938164 = 0.061836.

4. For Type A, the mean is (per policy)

and

V ar (X) = q (1 q) 2 + q = 0.3525.

62

ACTL3162 General Insurance Techniques Exercises

100

! P100 !

1 i=1 XAi 50 100 (0.15)

X

P XAi > 0.5 = P > p

100 i=1 0.3525 (100)

P100 !

i=1 XAi 50 100 (0.15)

= P > p

0.3525 (100)

= P (Z > 5.8951) = 1 P (Z 5.8951)

1 1 = 0.0

E[(S E[S]])3

S = X = = 2 1/2

S3

E[Y13 ]

= = 2 1/2 . (4.5)

(v)1/2 E[Y12 ]3/2

(v)E[Y12 ]3

=4 . (4.6)

E[Y13 ]2

(v)E[Y12 ]3 1 E[Y12 ]

c2 = 4 = c = 2 > 0. (4.7)

E[Y13 ]2 vE[Y12 ] E[Y13 ]

Finally substituting the solved and c into equation for the expectation gives k

(v)E[Y12 ]2

k = vE[Y1 ] 2 . (4.8)

E[Y13 ]

63

ACTL3162 General Insurance Techniques Exercises

Solution 4.3: [los31K, Exercise] Note that for compound Poisson moments, it can be derived

that

E[Z] = E[X]

V ar[Z] = E[X 2 ]

E [(X E[X])3 ]

Skew[Z] =

3

3

E[X ]

=

3

where X is the severity distribution and is the frequency severity. Plugging in these expres-

sions, we have that

E (S) = 1 = 6

and

1

V ar (S) = 2 = 12 =4

3

and

3 1 3

S = 3 / = 12 / 4 = 3/8.

4

(Note that the third moment of a U (a, b) distribution is 14 (a + b)(a2 + b2 ).

CLT:

P (S < 10) P [Z < (10 6) /2] = (2) = 0.977.

Translated Gamma: Using known expressions for the skewness of a Gamma distribution (skew-

ness = 2 ), we have that = 4/ 2 = 28 49 ;. Further, it is also known that =

2

= 83 ;. Finally,

using the expression in the lecture slides, we have that the shift x0 = 2

= 4 32 .Thus,

2 4 8

P (S < 10) G 10 4 ; 28 ; = 0.968.

3 9 3

NP approximation: Calculating s using the expressions in the lecture notes, we have that

P (S < 10) P (Z < 2) 97 8 = 0.968.

and

V ar (S) = 1000 12 0.01 0.99 + 1000 22 0.05 0.95 = 199.9

We know (y) = 0.95 for y = 1.645. So by letting P be equal to smallest premium income, it

should satisfy

P 110

P Z 0.95,

199.9

therefore the premium income must be P 110 + 1.645 199.9 = 133.258, and the loading

therefore has to be at least 23.258/110 = 21.14%.

64

ACTL3162 General Insurance Techniques Exercises

Solution 4.5: [los33K, Exercise] Let X1 be a claim of type 1, then P (X1 = 0) = 1 q1 , and

P (X1 = j) = q1 p1 (j) , for j = 1, 3. So

and

E X12 = 12 0.01 0.5 + 32 0.01 0.5 = 0.05

and

E X22 = 12 0.02 0.5 + 22 0.02 0.5 = 0.05

so that V ar (X2 ) = 0.0491. Then, calculating the expected value and variance of S:

and

E (S2 ) = 2000 0.03 = 60, V ar (S1 ) = 2000 0.0496 = 98.2

and thus

E (S1 + S2 ) = 20 + 60 = 80, V ar (S1 + S2 ) = 147.8.

This capital is p

E (S) + 1.645 V ar (S) = 99.999.

Solution 4.6: [los34, Exercise] Let S be based on j = log (1 qP j ) , bj > 0, for all j. Note

that qj < j since log (1 qj ) = qj + qj /2 + qj /3 + Since S = n1 Nj bj , so that

2 3

n

X n

X

E (S ) = j bj > qj bj = E (S) = E Se

1 1

and n n

X X

V ar (S ) = j b2j > qj b2j = V ar (S) > V ar Se .

1 1

which implies

P (Ni = 0) = 1 qi = P (Ii = 0)

P (Ni = 1) < P (Ii = 1)

P (Ni = 2, 3...) > P (Ii = 2, 3, ...) = 0.

Furthermore,

P (Ni = 1) = qi exp (qi ) < P (Ii = 1)

P (Ni = 2, 3...) > P (Ii = 2, 3, ...) = 0.

65

ACTL3162 General Insurance Techniques Exercises

X

E Se = ni qi bi = 25 (0.01 + 0.02 + 0.02 + 0.04) = 2 14

and X

V ar Se = ni qi (1 qi ) b2i = 3.6875

and X

e=ni qi (1 qi ) (1 2qi ) b3i / 3 = 0.906.1

= 2 14 ; 2 = 3.75; = 3 / (2 )3/2 = 6.75 3 = 0.93.

For Se :

= 4.871; = 1.149; x0 = 1.988.

For S :

= 4.630; = 1.111; x0 = 1.917.

#standardised s

z<-(s-Mom[1])/Mom[2]^.5

#Vector of results

aprob <- c()

# Normal Power #

################

aprob <- c(aprob,pnorm(s,Mom[1],Mom[2]^.5))

#Normal Power 2

temp <- sqrt(9/Mom[4]^2+6*z/Mom[4]+1)-3/Mom[4]

aprob <- c(aprob,pnorm(temp))

# Edgeworth #

#############

phi<-expression(1/sqrt(2*pi)*exp(-x^2/2))

dphi<-c(phi);

for (i in 2:6){dphi <- c(dphi,D(dphi[i-1],"x"))}

#Edgeworth 1

aprob <- c(aprob,pnorm(z)-Mom[4]/6*eval(dphi[3],list(x=z)))

#Edgeworth 2

aprob <- c(aprob,as.double(aprob[3])+Mom[5]/24*eval(dphi[4],list(x=z)))

#Edgeworth 3

aprob <- c(aprob,as.double(aprob[4])+Mom[4]^2/72*eval(dphi[6],list(x=z)))

# Translated Gamma #

####################

#parameters

tgbeta <- 2*Mom[2]/Mom[3]

tgalpha <- tgbeta^2*Mom[2]

x0 <- Mom[1]-tgalpha/tgbeta

#approximation

1

Correction to the skewness formula: made on 21-9-2004.

66

ACTL3162 General Insurance Techniques Exercises

row.names(approximat ions) <- paste("Pr[X",as.character(s),"]",sep="")

#print results

if(print==1){

cat("Approximations for Pr[X",as.character(s),"]\n",sep="")

print(approximations)

} # end if

Solution 4.9: [los41R, Exercise] For the solution code, refer to [los40R, Exercise]

[1] "Comparison of true probability and approximations for S(50)"

print(Proba,digits=5,print.gap=2)

True CLT Gamma NP

If N is Poisson 0.0091035 0.00250348 0.0103916 0.0107226

If N is Negative Binomial 0.0326376 0.01379840 0.0340177 0.0372348

If N is Binomial 0.0011910 0.00035914 0.0020282 0.0020194

2. Recall that we have computed the skewness of the distribution of S under the different

assumptions.

Var(.) 12.25 130.000 211.00 89.500

E[(.-E[.])^3] 6.00 1050.000 3534.00 354.750

gamma(.) 0.14 0.708 1.15 0.419

We can see that under the negative binomial and Poisson assumptions of N , the coeffi-

cients of skewness of S take value of 1.15 and 0.419. When approximating S, the trans-

lated gamma method produces the most accurate results as the method uses a gamma

distribution to approximate which itself is positively skewed. The normal power method

performs quite well, while the CLT method performs very poorly.

Under the binomial assumption of N , S has a coefficient of skewness of 0.419. In this

case, the normal power method performs the best as it uses the CLT idea which preserves

certain degree of symmetry and at the same time allows for a certain degree of skewness.

e n

Pr[N = n] n!

= e n1

= = 0+

Pr[N = n 1] (n1)!

n n

For negative binomial, we have

(n+r1)! r

Pr[N = n] n!(r1)!

p (1 p)n n+r1

= (n+r2)!

= (1 p)

Pr[N = n 1] pr (1 p)n1 n

(n1)!(r1)!

r1

= (1 p) 1 +

n

67

ACTL3162 General Insurance Techniques Exercises

For binomial, we have

m!

Pr[N = n] n!(mn)!

pn (1 p)mn mn+1 p

= m!

=

Pr[N = n 1] (n1)!(mn+1)!

pn1 (1 p)mn+1 n 1p

p m+1

= 1

1p n

s

1X

f (s) = hp (h) f (s h)

s h=1

1

f (s) = {.2f (s 1) + .8f (s 2) + 1.8f (s 3) + 3.2f (s 4)}

s

so that

f (0) = e2

f (1) = .2f (0) = .2e2

f (2) = .42f (0) = .42e2

and so on.

Note that f (0) will be a factor in all the f (x), x > 0. Thus, multiplying all the elements by

f (0) at the end only will be more efficient and will also prevent any rounding error (of the

evaluation of e2 ) to spread and amplify for large x.

Solution 4.12: [los26, Exercise] Note that iXi is compound Poisson with parameters (i =

i, pi (i) = 1). It follows from Theorem 12.4.1 that S is compound Poisson with parameters

= 1 + 2 + 3 = 6 and

1/6 x = 1,

2/6 x = 2,

p(x) =

3/6 x = 3,

0 elsewhere.

fS (0) = P (N = 0) = e1

68

ACTL3162 General Insurance Techniques Exercises

and

x

1X

fS (x) = (ax + bh) p (h) fS (x h)

x h=1

x

1X

= hp (h) fS (x h)

x h=1

1

= [0.2fS (x 1) + 0.4fS (x 2) + 0.6fS (x 3) + 1.6fS (x 4)]

x

where in the Poisson case, a = 0, b = . Therefore:

1

fS (2) = [0.2fS (1) + 0.4fS (0)] = 0.22e1

2

1

fS (3) = [0.2fS (2) + 0.4fS (1) + 0.6fS (0)] = 0.241e1

3

1

fS (4) = [0.2fS (3) + 0.4fS (2) + 0.6fS (1) + 1.6fS (0)] = 0.4641e1 .

4

...

Since Pr[X = 0] = 0

fS (0) = Pr[N = 0] = pr = 0.04.

We have then

min(4,s)

X s+j

fS (s) = 0.8 p(j)fS (s j), s = 1, 2, . . . .

j=1

s

and thus

fS (2) = 0.8 {3/2 0.2 0.0128 + 2 0.2 0.04} = 0.015872

fS (3) = = 0.01959936

fS (4) = = 0.03691315

...

x 1 j

n 1 X j e 1 n

p (x) = 1 (n + 1) 1 p (x j)

e j=1 x j!

x

X j 1 n

= (n + 1) 1 p (x j).

j=1

x j!

69

ACTL3162 General Insurance Techniques Exercises

We have then

pn (0) = (p(0))n = en

pn (1) = (n + 1 1)en = nen

n2

n n n+12 n

p (2) = e n+ = en

2 2 2

2

n3

n n n+13n 2n + 2 3 n n

p (3) = e + + = en .

3 2 3 2 3! 3!

# range of X

rX <- length(fX)-1

pmf <- c(fS0)

df <- c(fS0)

for(i in 1:vartype){ # i=s

temp <- sum((a+b*(1:min(i,rX))/i)*fX[(1:min(i,rX))+1]*pmf[i-(1:min(i,rX))+1])

pmf <- c(pmf, temp/(1-a*fX[1])) # we divide only at the end

df <- c(df,df[i]+pmf[i+1]) # the df...

} # end i loop

i <- vartype+1 # useful to know how many recursions we did for below

} else { # if we focus on the df

i <- 1 # since we use while, we need to create our own counter

while(df[i]<(1-vartype)){ # we can use while here

temp <- sum((a+b*(1:min(i,rX))/i)*fX[(1:min(i,rX))+1]*pmf[i-(1:min(i,rX))+1])

pmf <- c(pmf, temp/(1-a*fX[1]))

df <- c(df,df[i]+pmf[i+1])

i <- i+1 # increment the counter

# the number of recursions is simply i in this case (for below)

} # end while

}

# printing results

if(print==1) {

results <- data.frame(x=0:(i-1),fS=pmf,FS=df)

print(results)

} # end if

#returning results

array(c(pmf,df),c(i,2))

}

Note that we need to return an array now in order to be able to refer to the results as object[i,j]

with i for x and j for fS (j = 1) or FS (j = 2).

discretisation <-function(densityorcdf,type,h,m){

if(type==1) { #cdf

pmf <- c()

pmf <- c(densityorcdf(h/2))

for(i in 1:(m-1)) {

pmf <- c(pmf,densityorcdf(h*i+h/2)-densityorcdf(h*i-h/2))

}

pmf <- c(pmf,1-densityorcdf((m-.5)*h))

pmf } else { pmf <- c() #density

pmf <- c(as.double(integrate(densityorcdf,0,h/2)[1]))

for(i in 1:(m-1)) {

70

ACTL3162 General Insurance Techniques Exercises

}

pmf <- c(pmf,as.double(integrate(densityorcdf,(m-.5)*h,Inf)[1]))

pmf }# end else

} # end function

source("...09_IRM_tut_wk03_E12.R")

source("...09_IRM_tut_wk03_E13.R")

source("...09_IRM_tut_wk03_E14.R")

source("...09_IRM_tut_wk03_E15.R")

beta <-.2

lambda <- 20

s<-150

###########################################

# Calculate the approximate probabilities #

cgf <- expression(lambda*((beta/(beta-t))^alpha-1))

param <- list(lambda=20,alpha=1,beta=.2)

Moments <- CMom123Gam12(cgf,param)

Approx<-Approx(Moments,s,0)

##########################################

# Calculate the "Panjer" probability via #

# discretisation and Panjer #

cdf <- function(x){1-exp(-beta*x)}

pmfX <- discretisation(cdf,1,0.01,5000)

# we need thus s/h=150*100 recursions

pmfS <- Panjer(0,lambda,pmfX,exp(lambda*(pmfX[1]-1)),1,15000,0)

tprob <- pmfS[15001,2]

#################

# print results #

cat("Approximations for Pr[X",as.character(s),"]:\n",sep="")

print(results)

cat("\n","Deviation with Panjer probability:\n",sep="")

print(results-tprob)

The plot of fS (x) shows that the distribution of S tends to be normal when is big, but the

distribution is still slightly skewed (1 = 0.474), which is why the CLT approximation performs

badly.

71

ACTL3162 General Insurance Techniques Exercises

pmfS[, 1]

Index

Solution 4.18: [los40R, Exercise] This is the R-code for this question.

#Assignment 2008, Question 3

################################

# Assumptions and parameters #

################################

#Distribution of losses

p <- c(.2,.1,.05,.1,.05,.1,.05,.1,.05,.1,.1)

########

#this will be the array of the moments of X, S with Poisson, S with NB, S with B

Moments <- array (dim=c(6,4))

rownames(Moments) <- c("E[.]","E[.^2]","E[.^3]","Var(.)","E[(.-E[.])^3]","gamma(.)")

colnames(Moments) <- c("X "," S if N Poisson"," S if N Neg Bin"," S if N Binomial")

# Moments of X #

####################

Moments[2,1] <- sum(c(0:(length(p)-1))^2*p)

Moments[3,1] <- sum(c(0:(length(p)-1))^3*p)

Moments[4,1] <- Moments[2,1] - Moments[1,1]^2

Moments[5,1] <- Moments[3,1]-3*Moments[2,1]*Moments[1,1]+2*Moments[1,1]^3

Moments[6,1] <- Moments[5,1]/Moments[4,1]^(1.5)

# Initialization of parameters

P_lambda <- 4

NB_r <- 4

NB_p <- 0.5

B_n <- 8

B_p <- 0.5

#End of calculations

end <- 300

##############################################

# Distribution of S and moments of (S-d)+ #

##############################################

Output <- array(dim=c(3,end+1,7))

72

ACTL3162 General Insurance Techniques Exercises

# index 2: s+1

# index 3: "x","fs(x)","Fs(x)","d","E[(S-d)+]","E[(S-d)+^2]","Var[(S-d)+]"

# Preliminaries #

#####################

#Output[1,1,2] <- c(exp(P_lambda*(p[1]-1)))

#Output[2,1,2] <- c(p^NB_r)

#Output[3,1,2] <- c((1-B_p)^B_n )

Output[1,1,2] <- c( exp(P_lambda*(p[1]-1)))

Output[2,1,2] <- c( (NB_p/(1-(1-NB_p)*p[1]))^NB_r )

Output[3,1,2] <- c( ((1-B_p)+B_p*p[1])^B_n )

#Masses at 0 in Fs

Output[1,1,3] <- Output[1,1,2]

Output[2,1,3] <- Output[2,1,2]

Output[3,1,3] <- Output[3,1,2]

Output[1,1,5] = P_lambda*Moments[1,1] # expectation

Output[1,1,7] = P_lambda*Moments[2,1] # variance

Output[1,1,6] = Output[1,1,7] + Output[1,1,5]^2 # second moment around the origin

Output[2,1,5] = NB_r*(1-NB_p)/NB_p*Moments[1,1]

Output[2,1,7] = NB_r*(1-NB_p)/NB_p^2*Moments[1,1]^2 + NB_r*(1-NB_p)/NB_p*Moments[4,1]

Output[2,1,6] = Output[2,1,7] + Output[2,1,5]^2

Output[3,1,5] = B_n*B_p*Moments[1,1]

Output[3,1,7] = B_n*B_p*(1-B_p)*Moments[1,1]^2+B_n*B_p*Moments[4,1]

Output[3,1,6] = Output[3,1,7] + Output[3,1,5]^2

# a-b parameters

ab <- array(dim=c(3,2))

ab[1,1] <- 0

ab[1,2] <- P_lambda

ab[2,1] <- 1-NB_p

ab[2,2] <- (NB_r-1)*(1-NB_p)

ab[3,1] <- -B_p/(1-B_p)

ab[3,2] <- (B_n+1)*B_p/(1-B_p)

#pdf and cdf and SL premium moments for the three cases

for (i in 1:3)

{

Output[i,1,1] <- 0 # column x

Output[i,1,4] <- 0 # column d

for (s in 1:end)

{

sum[i] <- 0

for (h in 1:min(s,length(p)-1))

{

sum[i] <- sum[i] + (ab[i,1]+ab[i,2]*h/s)*p[h+1]*Output[i,s-h+1,2]

}

Output[i,s+1,1] <- s # label column x

Output[i,s+1,2] <- 1/(1-ab[i,1]*p[1])*sum[i] #fs

Output[i,s+1,3] <- Output[i,s,3] + Output[i,s+1,2] # Fs

Output[i,s+1,4] <- s # label column d

Output[i,s+1,5] <- Output[i,s,5] - 1 + Output[i,s,3] # E[(S-d)+]

Output[i,s+1,6] <- Output[i,s,6] - 2*Output[i,s,5]+1-Output[i,s,3]

Output[i,s+1,7] <- Output[i,s+1,6] - Output[i,s+1,5]^2

}

}

####################

# Moments of S #

####################

73

ACTL3162 General Insurance Techniques Exercises

# Remember: #

###############

##this will be the array of the moments of X, S with Poisson, S with NB, S with B

#Moments <- array (dim=c(6,4))

#rownames(Moments) <- c("E[.]","E[.^2]","E[.^3]","Var(.)","E[(.-E[.])^3]","gamma(.)")

#colnames(Moments) <- c("X "," S if N Poisson"," S if N Neg Bin"," S if N Binomial")

for (i in 2:4) {

Moments[1,i] <- sum(c(0:end)*c(Output[i-1,1:(end+1),2]))

Moments[2,i] <- sum(c(0:end)^2*c(Output[i-1,1:(end+1),2]))

Moments[3,i] <- sum(c(0:end)^3*c(Output[i-1,1:(end+1),2]))

Moments[4,i] <- sum((c(0:end)-Moments[1,i])^2*c(Output[i-1,1:(end+1),2]))

Moments[5,i] <- sum((c(0:end)-Moments[1,i])^3*c(Output[i-1,1:(end+1),2]))

Moments[6,i] <- Moments[5,i]/Moments[4,i]^(1.5)

}

################################

# Probability approximations #

################################

rownames(Proba) <- c("If N is Poisson","If N is Negative Binomial","If N is Binomial")

colnames(Proba) <- c("True","CLT","Gamma","NP")

ss <- 50

for (i in 1:3) {

Proba[i,1] <- 1-Output[i,ss+1,3]

Proba[i,2] <- 1-pnorm(ss,Moments[1,i+1],Moments[4,i+1]^.5)

Proba[i,3] <- 1-pgamma(ss-Moments[1,i+1]+2*Moments[4,i+1]^.5/Moments[6,i+1] ,

4/Moments[6,i+1]^2 , 2/Moments[6,i+1]/Moments[4,i+1]^.5)

Proba[i,4] <- 1-pnorm((9/Moments[6,i+1]^2+6*(ss-Moments[1,i+1])/Moments[4,i+1]^.5/Moments[6,i+1]+1)

^.5-3/Moments[6,i+1])

}

###################

# Print results #

###################

#######################################

#End of table

prints <- 80

rownames(Poisson) <- rep("",times=prints+1)

colnames(Poisson) <- c("x","fs(x)","Fs(x)","d","E[(S-d)+]","E[(S-d)+^2]","Var[(S-d)+]")

rownames(NegBin) <- rep("",times= prints +1)

colnames(NegBin) <- c("x","fs(x)","Fs(x)","d","E[(S-d)+]","E[(S-d)+^2]","Var[(S-d)+]")

rownames(Bin) <- rep("",times= prints +1)

colnames(Bin) <- c("x","fs(x)","Fs(x)","d","E[(S-d)+]","E[(S-d)+^2]","Var[(S-d)+]")

for (i in 1:7) {

for (j in 1:(prints+1)) {

Poisson[j,i] <- Output[1,j,i]

NegBin[j,i] <- Output[2,j,i]

Bin[j,i] <- Output[3,j,i]

}

}

print(Poisson,print.gap=3)

print(c("Distribution of S and stop-loss premiums if N is Negative Binomial"))

print(NegBin,print.gap=3)

print(c("Distribution of S and stop-loss premiums if N is Binomial"))

74

ACTL3162 General Insurance Techniques Exercises

print(Bin,print.gap=3)

# Moments #

###########

options(scipen=2)

print(Moments,digits=3)

# Probabilities #

#################

print(Proba,digits=5,print.gap=2)

# plots #

#########

mfrow=c(2,2)

plot(Poisson[,1],Poisson[,2],main="f(x) of S if N Poisson",xlab="",ylab="",col="1")

plot(NegBin[,1],NegBin[,2],main="f(x) of S if N NegBin",xlab="",ylab="",col="2")

plot(Bin[,1],Bin[,2],main="f(x) of S if N Bin",xlab="",ylab="",col="3")

plot(Poisson[,1],Poisson[,3],main="F(x) of S",xlab="",ylab="",col="1",type="l")

lines(NegBin[,1],NegBin[,3],col="2")

lines(Bin[,1],Bin[,3],col="3")

plot(Poisson[,4],Poisson[,5],type="l",col="1",main="Stop-loss Premium",xlab="d",ylab="E[(S-d)+]",)

lines(NegBin[,4],NegBin[,5],col="2")

lines(Bin[,4],Bin[,5],col="3")

legend(60,10,legend=c("Poi","NegBin","Bin"),lty=1,col=1:3)

print(c("Distribution of S if N is Poisson"))

[1] "Distribution of S if N is Poisson"

print(Poisson[,1:3],print.gap=3)

x fs(x) Fs(x)

0 0.040762203978 0.04076220

1 0.016304881591 0.05706709

2 0.011413417114 0.06848050

3 0.020000654752 0.08848116

4 0.016185312460 0.10466647

5 0.024547760130 0.12921423

6 0.021824859398 0.15103909

7 0.030071258549 0.18111035

8 0.028466826771 0.20957717

9 0.036711555213 0.24628873

10 0.044414433611 0.29070316

11 0.031582034174 0.32228520

12 0.032987043085 0.35527224

13 0.033859163507 0.38913140

14 0.034284589883 0.42341599

15 0.035304798458 0.45872079

16 0.034700813781 0.49342161

17 0.035645183751 0.52906679

18 0.033930905728 0.56299770

19 0.034555188644 0.59755288

20 0.032430919914 0.62998380

21 0.028716574671 0.65870038

22 0.027888692422 0.68658907

23 0.026623935971 0.71321301

24 0.025483357983 0.73869637

25 0.024071665116 0.76276803

[1] "Distribution of S if N is Negative Binomial"

print(NegBin[,1:3],print.gap=3)

x fs(x) Fs(x)

0 0.0952598689 0.09525987

1 0.0211688598 0.11642873

2 0.0135245493 0.12995328

3 0.0244356591 0.15438894

4 0.0177214790 0.17211042

5 0.0283404926 0.20045091

6 0.0225677187 0.22301863

75

ACTL3162 General Insurance Techniques Exercises

7 0.0329761903 0.25599482

8 0.0281639543 0.28415877

9 0.0384485051 0.32260728

10 0.0452089616 0.36781624

11 0.0266490320 0.39446527

12 0.0275746285 0.42203990

13 0.0278652951 0.44990519

14 0.0279380518 0.47784325

15 0.0284733228 0.50631657

16 0.0276653689 0.53398194

17 0.0282848455 0.56226678

18 0.0265446534 0.58881144

19 0.0270714614 0.61588290

20 0.0250558092 0.64093871

21 0.0218626373 0.66280134

22 0.0213481731 0.68414952

23 0.0205224539 0.70467197

24 0.0198092739 0.72448125

25 0.0189021953 0.74338344

print(c("Distribution of S if N is Binomial"))

[1] "Distribution of S if N is Binomial"

print(Bin[,1:3],print.gap=3)

x fs(x) Fs(x)

0 1.679616e-02 0.01679616

1 1.119744e-02 0.02799360

2 8.864640e-03 0.03685824

3 1.500768e-02 0.05186592

4 1.382022e-02 0.06568614

5 1.988766e-02 0.08557380

6 1.986838e-02 0.10544218

7 2.599662e-02 0.13143879

8 2.717911e-02 0.15861790

9 3.350957e-02 0.19212747

10 4.153728e-02 0.23366474

11 3.468674e-02 0.26835149

12 3.666887e-02 0.30502036

13 3.836575e-02 0.34338611

14 3.931281e-02 0.38269892

15 4.096584e-02 0.42366476

16 4.079371e-02 0.46445847

17 4.210595e-02 0.50656442

18 4.069388e-02 0.54725830

19 4.134746e-02 0.58860576

20 3.935274e-02 0.62795850

21 3.533203e-02 0.66329053

22 3.406891e-02 0.69735943

23 3.222769e-02 0.72958712

24 3.050720e-02 0.76009432

25 2.843743e-02 0.78853176

options(scipen=2)

print(c("Moments of the 4 distributions"))

[1] "Moments of the 4 distributions"

print(Moments,digits=3)

X S if N Poisson S if N Neg Bin S if N Binomial

E[.] 4.50 18.000 18.00 18.000

E[.^2] 32.50 454.000 535.00 413.500

E[.^3] 262.50 13902.000 20760.00 11019.750

Var(.) 12.25 130.000 211.00 89.500

E[(.-E[.])^3] 6.00 1050.000 3534.00 354.750

gamma(.) 0.14 0.708 1.15 0.419

76

ACTL3162 General Insurance Techniques Exercises

N1 Poisson(1 = 4),

N2 Neg Bin(r2 = 4, p2 = 0.5), and

N3 Bin(n3 = 8, p3 = 0.5).

The random variables N1 , N2 and N3 all have the expected value of 4. But their variances

are different. N1 has variance of 4, N2 has variance of 8 and N3 has variance of 2. Recall

that N in the collective risk model is the random variable that represents the number of

claims in a portfolio. The variability of this random variable will play a vital role in the

distribution of S. We will illustrate this using the following graphs.

Binomial Poisson

0.04

0.04

0.03

0.03

Poisson[1:81, 2]

0.02

0.02

0.01

0.01

0.00

0.00

0 20 40 60 80 0 20 40 60 80

Bin[1:81,

Negative 1]

Binomial Poisson[1:81,

Distribution 1]

functions

1.0

0.08

0.8

Poisson[1:81, 3]

0.06

0.6

0.04

0.4

0.02

0.2

0.00

0 20 40 60 80 0 20 40 60 80

The first three graphs represent the distribution of S under each assumption of Ni , i =

1, 2, 3. Notice that under the negative binomial assumption, S has a big tail. Under the

Poisson assumption, the tail is smaller. Under the binomial assumption, the tail is the

smallest. This is caused by the different values of variance of the random variables Ni

under different assumptions. Under the negative binomial assumption, N has the largest

variance. Therefore this causes a large variability in the entire portfolio which is reflected

by the big tail.

77

ACTL3162 General Insurance Techniques Exercises

This idea is reinforced by the fourth graph, which is the cumulative distribution function

under each assumptions of Ni , i = 1, 2, 3. We notice that, under the negative binomial

assumption (red line), more probabilities are assigned to the larger values of x. In com-

parison, for the binomial assumption, more probabilities are assigned to small values of

x.

78

Module 5

Exercise 5.1: [sur2, Solution] An insurance company has a portfolio of policies with the

following characteristics:

x p (x)

2 0.2

4 0.5

6 0.3

Determine the smallest relative security loading the insurer can choose so that it is certain

that ruin does not occur at time 1.

Exercise 5.2: [sur3, Solution] Show that all these expressions are equivalent (to determine R)

+ r = mX (r) (2)

Z

[erx (1 + )] [1 P (x)]dx = 0 (3)

0

er = E[erS(1) ] (4)

where p1 = E(X).

79

ACTL3162 General Insurance Techniques Exercises

E eRC(t) = eRc0 .

lim R

0

and

lim R

1

erx > 1 + rx + (rx)2 , r > 0, x > 0 (1)

2

to show that

2p1

R< ,

p2

Exercise 5.6: [sur7, Solution] Suppose that claims form a compound Poisson process, with

= 1 and p (x) = 1 for 0 < x < 1.

Premiums are received continuously at the rate of c = 1.

Find the adjustment coefficient if proportional reinsurance is purchased with a retention = 0.5

and with reinsurance loading equal to 100%.

Exercise 5.7: [sur8, Solution] An insurance company has aggregate claims that have a com-

pound Poisson distribution with:

= 2;

p (x) = 1/2, 0 x 2; and

premium collection rate is 6.

The insurer buys a proportional reinsurance that reimburses 20% of each individual claim.

The adjustment coefficient with reinsurance is 1.75. Determine the reinsurance loading.

1. Find R if c = 3.

2. Plot R for = i/100, i = 1, 2, . . . , 100.

Exercise 5.9: [sur10R, Solution][R ] Consider Exercise 5.6 above. Assume now that = 0.60

(where this is the loading of the insurer). Find the optimal that will minimise the probability

of ruin and plot R for = 0.5 + i/100, i = 1, 2, . . . , 49.

Exercise 5.10: [sur11, Solution] Consider a Brownian motion process with drift and volatil-

ity 2 . Using the moment generating function to show that the Brownian motion can be

approximated by a shifted Poisson process { N (t) t, t 0}.

[Hint: you need to determine (, , ) and c(, , )]

80

ACTL3162 General Insurance Techniques Exercises

motion by a shifted Poisson process as seen in Exercise 5.10. Plot your process for a number

of jumps equal to three times (approximately three units of time).

If you set the seed for your exponential pseudo-random variables to 12345 (use set.seed(12345))

and if = 0.03, = 3 and = 6, you should get the following graph:

20

15

10

W(t)

5

0

Exercise 5.12: [cop1, Solution] Suppose (X1 , X2 ) has a bivariate distribution function de-

scribed by

( 1/ )

1 1

P (X1 x1 , X2 x2 ) = exp log + log

1 ex1 1 ex2

for some 1.

2. Express the bivariate joint distribution as a copula function and show that this copula

belongs to the class of Archimedean copulas. Identify the generator.

3. For what value of do you have the case of independence?

Exercise 5.13: [cop2, Solution] Let Y1 and Y2 be two (underlying) lifetimes with distribution

functions H1 and H2 . Suppose there exists independent rv Z Exp() that represents the

time until a common disaster. Assume both lives are subject to the same disaster, so that the

age-at-death random variables X1 and X2 are given by

X1 = min (Y1 , Z) and X2 = min (Y2 , Z) .

Additionally assume that Y1 Exp(1 ) and Y2 Exp(2 ).

81

ACTL3162 General Insurance Techniques Exercises

2. Derive the joint distribution function of (X1 , X2 ). Express your answer in the form of a

copula. This form of the copula is sometimes referred to as the Marshall-Olkin copula.

Exercise 5.14: [cop3, Solution] Let (X1 , X2 ) be a bivariate random vector with distribution

function described by a copula as

P (X1 x1 , X2 x2 ) = C (P (X1 x1 ) , P (X2 x2 ))

2. Express the distribution function of max (X1 , X2 ) in terms of the copula.

Exercise 5.15: [cop4, Solution] Consider the case of the bivariate Normal copula written as

C (u1 , u2 ) = 1 (u1 ) , 1 (u2 )

where denotes the joint distribution function of a bivariate standard Normal as given by

Z x2 Z x1

1 s2 + t2 2st

1

(x1 , x2 ) = p exp dsdt.

2 1 2 2 (1 2 )

1. Prove that we can re-write the equivalent form of the bivariate Normal copula as

Z u1 !

1 (u2 ) 1 (z)

C (u1 , u2 ) = p dz.

0 1 2

2. Prove that the bivariate Normal copula generates the bivariate Normal distribution if and

only if the marginals are standard Normal.

3. Find an expression for the copula density

2 C (u1 , u2 )

c (u1 , u2 ) =

u1 u2

and explain the significance of the copula density.

Exercise 5.16: [cop5R, Solution][R ] The loss and ALAE recorded for each of 24 claims are

provided in the table below:

Loss ALAE Loss ALAE

2,000 3,043 12,500 165

2,500 415 14,000 175

2,500 4,940 14,750 28,217

4,500 395 15,000 2,072

5,000 25 17,500 6,328

5,750 34,474 19,833 212

7,000 50 30,000 2,172

7,000 10,593 33,033 7,845

7,500 50 44,887 2,178

9,000 406 62,500 12,251

10,000 1,174 210,000 7,357

82

ACTL3162 General Insurance Techniques Exercises

Assume that loss has a Pareto distribution with density of the form

fLoss (x) = for x > 0

(x + )+1

and that ALAE has a Gamma distribution with density of the form

1 x

fALAE (x) = x e for x > 0.

()

1. Write an R code to estimate the parameters, together with their standard errors, assuming

loss and ALAE are independent random variables.

2. Re-write your R code to estimate the parameters, together again with their associated

standard errors, assuming this time that loss and ALAE follow a Frank copula with the

form !

1 eu1 1 eu2 1

C (u1 , u2 ) = log 1 + for 6= 0.

e 1

3. Based on the results of your model estimation, is there evidence to support that loss and

ALAE are not independent?

The following algorithm generates random variates (u, v) from a Farlie-Gumbel-Morgenstern

distribution with parameter :

p

Set a = (1 2u) 1; b = a2 4(a + 1)t

Set v = 2t/(b a)

In addition, the following algorithm generates random variates (u, v) from a Plackett distribu-

tion with parameter :

2 2

Set

pa = c(1 t); b = + a( 1) ; c = 2a(u + 1 u) + (1 2a); and d =

2

+ 4au(1 u)(1 ) ;

Set v = [c (1 2t)d]/2b

It can also be shown that Spearmans for a member of the Plackett family of copulas with

parameter > 0 is given by

+1 2

= ln .

1 ( 1)2

Generate 1000 random variates (u, v) from a Plackett distribution with parameter = 1. Use

R to do this, carefully documenting your code. Calculate the corresponding pair of random

variables (x, y) for a bivariate Pareto distribution; that is, with two Pareto marginals FX (x)

83

ACTL3162 General Insurance Techniques Exercises

and FY (y), each with parameters (2, 200). Repeat this exercise for parameters = 0.36 and

= 2.7. Plot your six outcomes on one page (using R), with the first column containing the

(u, v) results, and the second column containing the (x, y) results. The rows should contain

ascending values of .

Repeat the task outlined in the last paragraph, except now for random variates (u, v) from a

Farlie-Gumbel-Morgenstern distribution; the value of used should reflect an equivalent value

of Spearmans to the value of of the previous paragraph.

Comment on your results.

5.5 Solutions

Solution 5.1: [sur2, Exercise] Note that this surplus process is not a Cramer-Lundberg process

since the number of claims is not random.

The surplus process can still be expressed as C (t) = c0 + t S (t) where u = 1 and the

premium rate = E (X) (1 + ) = 4.2 (1 + ). Thus,

= P (1 + 4.2 (1 + ) S (1) < 0)

= P (S (1) > 5.2 + 4.2) .

Solution 5.2: [sur3, Exercise] Recall that the adjustment coefficient R is the positive solution

to:

1 + (1 + ) p1 R = mX (R) .

If we multiply both sides by , we have

+ (1 + ) p1 R = mX (R)

+ R = mX (R) .

1 + (1 + ) E (X) R = E eRX

so that Z Rx

e 1 (1 + ) Rx dP (x) = 0.

0

This implies

Z Z x Z Z

R eRy (1 + ) R dydP (x) = R eRy (1 + ) R

dP (x) dy

0 0 0 y

Z

(1 P (y)) R eRy (1 + ) R dy

=

0

= 0.

84

ACTL3162 General Insurance Techniques Exercises

R = [mX (R) 1]

eR = exp { [mX (R) 1]}

eR = mS(1) (R) = E eRS(1)

E eR(S(1)) = 1

1

= log mS(1) (R)

R

which confirms (4).

E eRC(t) = E eR[c0 +tS(t)] = eRc0 eRt E eRS(t) = eRc0 eRt E etRS(1)

E eRC(t) = eRc0 .

Solution 5.4: [sur5A, Exercise] Think of the graph of 1 + (1 + )p1 r against mX (r) seen in

lecture.

As 0, the slope of the line tends to p1 , which is exactly m0X (0). There is then only one

root, R = 0, and thus (u) = 1, u 0.

The case is less clear. As , the intersection of 1 + (1 + )p1 r with mX (r) will be

at the limit of mX (r) as r (remember that m0X (r) > 0, r > 0). But, the mgf of X is not

always defined for r [0, ). For instance, if X is exponential(), R as . On the

other hand, if X is inverse Gaussian, there will be no other root than the trivial 0 as

(see A, example 13.4.3.e on page 412).

Solution 5.5: [sur6A, Exercise] Note that the expression on the right hand side of (1) is the

beginning of Taylors

expansion of the exponential on the left hand side. Now remember that

rX

mX (r) = E e and thus

1

1 + (1 + )p1 R = E eRX > 1 + Rp1 + R2 p2 .

2

Subtracting 1 on both sides, dividing by R and reorganising yields to the desired expression.

Z 1

h = 2 h (x) p (x) dx = 1

0

and therefore the adjustment coefficient Rh satisfies

Z 1

eRh 1

1 + Rh = eRh ()x dx = .

0 Rh

In the absence of reinsurance ( = 1), the non-trivial solution of this equation can be shown

to be R = 1.793. But if = 0.5, it is clear that

Rh = 1.793/ (1 0.5) = 3.587.

85

ACTL3162 General Insurance Techniques Exercises

Solution 5.7: [sur8, Exercise] Since reinsurance is proportional, h (x) = (1 )x = 0.2x and

reinsurance premium is

Z 2

h = (1 + ) h (x) p (x) dx

0

= 0.4 (1 + )

Z 2

+ ( h ) Rh = eRh (xh(x)) p (x) dx

Z 02

2 + [6 0.4 (1 + )] (1.75) = 2 e1.75(0.8x) 0.5dx

0

Solution 5.8: [sur9R, Exercise] We solve for the positive R which is the solution to + R =

mX (R). The m.g.f. is given by

1

1 + et + e2t + e4t

mX (t) =

4

so that the equation to solve is

1

1 + eR + e2R + e4R = 0.

1 + R

4

0.4

0.3

0.2

R

0.1

0.0

theta

86

ACTL3162 General Insurance Techniques Exercises

eqR <- function(r,theta){

return(1+(1+theta)*7/4*r-(1+exp(r)+exp(2*r)+exp(4*r))/4)

}

# create a function for finding the root of eqR for given theta

fR <- function(x){

uniroot(eqR,lower=0.001,upper=2,theta=x)$root

}

# part a

R <- fR(5/7)

print(R)

# part b

plot <- c()

for(i in 1:100){plot <- c(plot,fR(i/100))} # the y values

plot(1:100/100,plot,xlab="theta",ylab="R")

1.40

1.35

1.30

1.25

R

1.20

1.15

1.10

alpha

87

ACTL3162 General Insurance Techniques Exercises

return( 1+(alpha-.2)*r-(exp(alpha*r)-1)/alpha/r )

}

# create a function for finding the root of eqR for given theta

fR <- function(x){

uniroot(eqR,lower=0.001,upper=100,alpha=x)$root

}

R <- optimise(fR,interval=c(0.5,0.999),maximum=TRUE)

print(R)

#plotting

plot <- c()

for(i in 1:49) plot <- c(plot,fR(.5+i/100))

plot(51:99/100,plot,xlab="alpha",ylab="R")

Solution 5.10: [sur11, Exercise] Consider the shifted Poisson process { N (t) t, t 0} and

we will match the mean and variance of the shifted Poisson process with the Brownian motion

(1 unit of time), i.e.

= and 2 = 2 . (5.1)

Solving for and yields

2 2

= and = . (5.2)

2

Next we obtain the moment generating function of process { N (t) t}

k

E ek( N (t)t) = etk E ek( N (t)) = etk et(e 1) .

(5.3)

(k )2 (k )3

ek 1 = k + + + ... (5.4)

2! 3!

If we substitute the expansion back into (5.3) and make change of and c using (5.2), we obtain

2 2 (k )2 (k )3

tk+ 2 t k + 2! + 3! +...

E ek( N (t)t) = e

2 2 2

tk+tk+ tk+ 2 tk2 +o( )

= e . (5.5)

2

(t)k+ 2 t k2

E e

k( N (t)t)

=e , (5.6)

which is the moment generation function of a Brownian motion process with drift and volatil-

ity 2 .

88

ACTL3162 General Insurance Techniques Exercises

# = total net loss for constant claims tau

# X(t) = tau*N(t) - ct

# E[X(t)]=(tau*lambda-c) t

# Var(X(t))= tau^2 lambda t

###

# we choose lambda and c such that

# mu = tau lambda - c

# sigma^2 = tau^2 * lambda

###

# we have then

# lambda = sigma^2/tau^2

# c = sigma^2/tau - mu

# and we will let tau -> 0

#initialising parameters

mu <- 3

sigma <- 6

tau <- .03

c <- sigma^2/tau - mu

# number of jumps

num <- round(3*lambda)

set.seed(12345)

times <- rexp(num,lambda)

# calculate process

Brownian <- array(c(rep(0,2*(num+1))),c(num+1,2))

for(i in 1:num){

Brownian[i+1,1] <- Brownian[i,1] + times[i]

Brownian[i+1,2] <- tau*i-c*Brownian[i+1,1]

} #end of for loop

# plot Brownian

colnames(Brownian)<-c("t","W(t)")

plot(Brownian,type="l",lwd="0.1",main="Approximation of a Brownian Motion")

F2 (x2 ) = F (, x2 ) = 1 ex2 .

89

ACTL3162 General Insurance Techniques Exercises

2. Let u1 = F1 (x1 ) and u2 = F2 (x2 ). We find that we can write the bivariate joint distri-

bution function as

n o

1 1/

P (X1 x1 , X2 x2 ) = C (u1 , u2 ) = exp log u11 + log u2

n o

1/

= exp [( log u1 ) + ( log u2 ) ] .

This is the required copula function and we can indeed re-write this as

(u) = ( log u) .

= exp {log u1 + log u2 } = u1 u2

Solution 5.13: [cop2, Exercise] Denote the marginals of X1 and X2 by F1 and F2 , respectively.

= 1 P (Y1 > x1 and Z > x1 ) = 1 P (Y1 > x1 ) P (Z > x1 )

= 1 exp (1 x1 ) exp (x1 ) = 1 exp ( ( + 1 ) x1 ) .

= 1 P (Y2 > x2 and Z > x2 ) = 1 P (Y2 > x2 ) P (Z > x2 )

= 1 exp (2 x2 ) exp (x2 ) = 1 exp ( ( + 2 ) x2 ) .

F (x1 , x2 ) = P (X1 x1 , X2 x2 )

= P (X1 x1 ) + P (X2 x2 ) + P (X1 > x1 , X2 > x2 ) 1

where

= P (Y1 > x1 , Y2 > x2 , Z > max (x1 , x2 ))

= P (Y1 > x1 ) P (Y2 > x2 ) P (Z > max (x1 , x2 ))

= exp (1 x1 ) exp (2 x2 ) exp ( max (x1 , x2 )) .

+ exp (1 x1 ) exp (2 x2 ) min (exp (x1 ) , exp (x2 )) .

90

ACTL3162 General Insurance Techniques Exercises

= exp ( ( + 1 ) x1 ) exp ( ( + 2 ) x2 ) exp ( (x1 + x2 ))

exp ( max (x1 , x2 ))

= exp ( ( + 1 ) x1 ) exp ( ( + 2 ) x2 )

min (exp (x1 ) , exp (x2 ))

= [P (X1 > x1 )] [P (X2 > x2 )]

+ +

min [P (X1 > x1 )] 1 , [P (X2 > x2 )] 2

1

!

[P (X2 > x2 )] [P (X1 > x1 )] 1 + ,

= min 1

.

[P (X1 > x1 )] [P (X2 > x2 )] 2 +

We find that the required copula for the Marshall-Olkin bivariate exponential distribution

has the form

C (u1 , u2 ) = u1 + u2 1

1 1

+ min (1 u2 ) (1 u1 ) 1 + , (1 u1 ) (1 u2 ) 2 +

and

u2 = P (X2 x2 ) = 1 P (X2 > x2 ) .

1. Note that

= 1 P (X1 > x, X2 > x)

= 1 [1 P (X1 x) P (X2 x) + P (X1 x, X2 x)]

= P (X1 x) + P (X2 x) P (X1 x, X2 x)

= P (X1 x) + P (X2 x) C (P (X1 x) , P (X2 x)) .

= C (P (X1 x) , P (X2 x)) .

1

Corrections made as at 1 Nov 2006. Thanks to Eugenia.

91

ACTL3162 General Insurance Techniques Exercises

Z 1 (u1 ) Z 1 (u2 )

1 s2 + t2 2st

1

C (u1 , u2 ) = p exp dsdt

2 1 2 2 (1 2 )

Z 1 (u1 ) Z 1 (u2 ) !

1 1 (s t)2 + t2 2 t2

= p exp dsdt

2 1 2 2 (1 2 )

!2

Z 1 (u1 ) Z 1 (u2 )

1 2 1 1 s t

= e 2 t p exp p dsdt

2 1 2 2 1 2

s t ds

Now apply a change of variable w = p so that dw = p and

1 2 1 2

1 1 t2 ( (u2 )t)/ 1 1

Z x1 Z 1 2

1 2

C (u1 , u2 ) = e 2 exp w dwdt

2 2 2

Z x1

1 1 2

p

e 2 t 1 (u2 ) t / 1 2 dt

=

2

Now applying the transformation t = 1 (z) so that

1 1 2

dz = e 2 t dt,

2

we find that

!

u1 1 1

(u2 )

Z

(z)

C (u1 , u2 ) = p dz

0 1 2

!

u1 1 1

(u2 )

Z

(z)

= p dz.

0 1 2

Indeed this is a nice result!.

2. One direction is pretty obvious: if the marginals are standard Normal, then the bivariate

Normal copula generates a bivariate Normal distribution. Consider that

Z x1 Z x2

1 s2 + t2 2st

1

C (F1 (x1 ) , F2 (x2 )) = p exp dsdt

2 1 2 2 (1 2 )

which is the Normal copula with Normal marginals if and only if

1 (F1 (x1 )) = x1 and 1 (F2 (x2 )) = x2

whenever F1 = F2 = , the Normal marginals.

3. Using the canonical representation of the bivariate Normal with standard Normal marginals,

we find that2

1 x21 + x22 2x1 x2

1

f (x1 , x2 ) = p exp

2 1 2 2 (1 2 )

2

1 x

= c ( (x1 ) , (x2 )) exp 1

2 2

2

1 x

exp 2

2 2

2

Corrections made as at 3 Sep 2010. Thanks to Alex, Kieran.

92

ACTL3162 General Insurance Techniques Exercises

so that

x2 +x22 2x1 x2

1

exp 12 1 (1 2)

2 12

c ( (x1 ) , (x2 )) = 2 2

x x

1 exp 1 1 exp 2

2 2 2 2

2 2 2 2

1 x1 + x2 2x1 x2

= p exp .

1 2 2 (1 2 )

!

1 2 1 (u1 )2 + 2 1 (u2 )2 21 (u1 ) 1 (u2 )

c (u1 , u2 ) = p exp .

1 2 2 (1 2 )

To interpret, the copula density distorts the independence to induce the actual dependence

structure.

Solution 5.16: [cop5R, Exercise] R code for optimization of log-likelihood both the case of

independence and the case of Frank copula.

1. The following is the R code for maximizing the log-likelihood for the case of independence.

#

# read the data file

source("DataSummStats.R")

source("dPareto.R")

source("pPareto.R")

attach(Loss.ALAE)

DataSummStats(ALAE)

"neg.loglik" <- function(parm,x1,x2)

{

p.lambda <- parm[1]

p.theta <- parm[2]

g.alpha <- parm[3]

g.beta <- parm[4]

f1 <- dPareto(x1,alpha=p.theta,x0=p.lambda)

f2 <- dgamma(x2,shape=g.alpha,rate=1/g.beta)

93

ACTL3162 General Insurance Techniques Exercises

return(sum(-log.ll))

}

init.est <- c(36000,2.6,0.75,804)

parm.hat <- fit.Indep\$par

loglik.Indep <- -fit.Indep\$value

AIC.Indep <- (-2*loglik.Indep+2*length(parm.hat))/length(Loss)

print(loglik.Indep)

print(fit.Indep)

print(AIC.Indep)

library(nlme)

negll.Indep.Hess <- fdHess(parm.hat, neg.loglik, x1=Loss,x2=ALAE)

inv.Indep.Hess <- solve(negll.Indep.Hess\$Hessian)

parm.se <- sqrt(diag(inv.Indep.Hess))

output <- cbind(parm.hat,parm.se) output <- round(output,digits=3)

rownames(output)<- c("lambda","theta","alpha","beta")

colnames(output)<- c("estimate","std error") print(output)

The results of the estimation, together with the standard errors are as follows:

lambda 32536.132 26066.697

theta 2.463 1.490

alpha 0.443 0.105

beta 12002.412 4659.556

2. The following is the R code for maximizing the log-likelihood for the case of the Frank

copula.

#

# read the data file

source("DataSummStats.R")

source("dPareto.R")

source("pPareto.R")

attach(Loss.ALAE)

DataSummStats(ALAE)

"C12.Frank" <- function(u1,u2,d.alpha) {

tmp1 <- exp(d.alpha)-1

tmp2 <- exp(d.alpha*(u1+u2))

tmp3 <- (exp(d.alpha*u1)-1)*(exp(d.alpha*u2)-1)

num <- d.alpha*tmp1*tmp2

den <- (tmp1+tmp3)^2

94

ACTL3162 General Insurance Techniques Exercises

return(num/den)

} "neg.loglik" <- function(parm,x1,x2) {

p.lambda <- parm[1]

p.theta <- parm[2]

g.alpha <- parm[3]

g.beta <- parm[4]

f.alpha <- parm[5]

f1 <- dPareto(x1,alpha=p.theta,x0=p.lambda)

f2 <- dgamma(x2,shape=g.alpha,rate=1/g.beta)

u1 <- pPareto(x1,alpha=p.theta,x0=p.lambda)

u2 <- pgamma(x2,shape=g.alpha,rate=1/g.beta)

f12 <- C12.Frank(u1,u2,f.alpha)

log.ll <- log(f1)+log(f2)+log(f12)

return(sum(-log.ll))

} init.est <- c(36000,2.6,0.75,804,-2.1)

parm.hat <- fit.Frank\$par

loglik.Frank <- -fit.Frank\$value

AIC.Frank <- (-2*loglik.Frank+2*length(parm.hat))/length(Loss)

print(loglik.Frank)

print(fit.Frank)

print(AIC.Frank)

library(nlme)

negll.Frank.Hess <- fdHess(parm.hat, neg.loglik, x1=Loss,x2=ALAE)

inv.Frank.Hess <- solve(negll.Frank.Hess\$Hessian)

parm.se <- sqrt(diag(inv.Frank.Hess))

output <- cbind(parm.hat,parm.se)

output <- round(output,digits=3)

rownames(output)<- c("lambda","theta","alpha","beta","delta")

colnames(output)<- c("estimate","std error") print(output)

The results of the estimation, together with the standard errors are as follows:

lambda 31682.240 23288.123

theta 2.293 1.287

alpha 0.432 0.105

beta 11699.070 4474.634

delta -2.217 1.420

H0 : = 0

against the alternative

H0 : 6= 0

based on the Frank copula model. This is because when = 0 in the Frank copula, we

get the independence copula (verify this by taking limit!). The test statistic is

b 0 2.217 0

=

= 1.56 < 2.

s.e.

b 1.420

95

ACTL3162 General Insurance Techniques Exercises

So therefore, we cannot reject the null hypothesis that Loss and ALAE are independent.

There is no strong evidence to support that they are not independent. Indeed, if we

examine the scatterplot between Loss and ALAE, this also helps to confirm this because

we see no evidence of any particular dependencies between Loss and ALAE.

10

8

log(ALAE)

6

4

8 9 10 11 12

log(Loss)

For FGM copulas with parameter , Spearmans rho is given by = /3. We have then

parameter

Plackett FGM Type of dependence

0.36 -0.987 -0.329 negative

1.00 0.000 0.000 independence

2.70 0.962 0.321 positive

In fact, the Plackett copulas allow for stronger negative and positive dependences than

the FGM copulas, whose parameter must lie within [1, 1], whereas the Plackett can

take any positive real value. Note that

2 1 2 2 2 ln 1

= ln = .

( 1)2 ( 1)2 ( 1)2

When = 1, this expression is indeterminate. Using de lHospital rule (twice), we get

d2

2 2 ln 1 d2

(2 2 ln 1) 2 2/

lim = lim d2

= lim = 0,

1 ( 1)2 1

d2

( 1)2 1 2

which could be expected since a Plackett copula with = 1 is the independence copula.

96

ACTL3162 General Insurance Techniques Exercises

Here is a possible code:

# number of couples

n = 1000

# parameters

theta <- c(.36,1,2.7)

pty="s", # to have squares and not rectanges

mar=rep(3,4), # 4 margins around plots

font.lab=3, # to have italics in the labels

mgp=c(2,.5,0)) # margins around title and labels

for (i in 1:3) {

u <- runif(n) # our first independent variables set us

t <- runif(n) # our second independent variables set ts

v <- c(rep(0,n)) # initialise the vector for the vs

#algorithm to create dependent vs out of the ts with Plackett

a <- t*(1-t)

b <- theta[i]+a*(theta[i]-1)^2

c <- 2*a*(u*theta[i]^2+1-u)+theta[i]*(1-2*a)

d <- sqrt(theta[i])*sqrt(theta[i]+4*a*u*(1-u)*(1-theta[i])^2)

v <- (c-(1-2*t)*d)/2/b

# mapping marginals into Paretos

x <- 200*((1-u)^(-1/2)-1)

y <- 200*((1-v)^(-1/2)-1)

# plots for that theta

plot(cbind(u,v),

main=paste("Plot of v against u, case theta=",theta[i]),

pch=20) # for small dots

plot(cbind(x,y),

main=paste("Plot of y against x, case theta=",theta[i]),

pch=20,

xlim=c(0,800),ylim=c(0,800)) #up to the 96% quantile

} # end i loop

# number of couples

n = 1000

# parameters

theta <- c(-.987,0,.962)

pty="s", # to have squares and not rectanges

mar=rep(3,4), # 4 margins around plots

97

ACTL3162 General Insurance Techniques Exercises

mgp=c(2,.5,0)) # margins around title and labels

for (i in 1:3) {

u <- runif(n) # our first independent variables set us

t <- runif(n) # our second independent variables set ts

v <- c(rep(0,n)) # initialise the vector for the vs

#algorithm to create dependent vs out of the ts with FGM

a <- 1+theta[i]*(1-2*u)

b <- sqrt(a^2-4*(a-1)*t)

v <-2*t/(b+a)

# mapping marginals into Paretos

x <- 200*((1-u)^(-1/2)-1)

y <- 200*((1-v)^(-1/2)-1)

# plots for that theta

plot(cbind(u,v),

main=paste("Plot of v against u, case theta=",theta[i]),

pch=20) # for small dots

plot(cbind(x,y),

main=paste("Plot of y against x, case theta=",theta[i]),

pch=20,

xlim=c(0,800),ylim=c(0,800)) #up to the 96% quantile

} # end i loop

Scatterplots for the Plackett variates:

98

ACTL3162 General Insurance Techniques Exercises

Plot of v against u, case theta= 0.36 Plot of y against x, case theta= 0.36

800

1.0

0.8

600

0.6

400

v

y

0.4

200

0.2

0.0

0

0.0 0.2 0.4 0.6 0.8 1.0 0 200 400 600 800

u x

800

1.0

0.8

600

0.6

400

v

y

0.4

200

0.2

0.0

0.0 0.2 0.4 0.6 0.8 1.0 0 200 400 600 800

u x

Plot of v against u, case theta= 2.7 Plot of y against x, case theta= 2.7

800

1.0

0.8

600

0.6

400

v

y

0.4

200

0.2

0.0

0.0 0.2 0.4 0.6 0.8 1.0 0 200 400 600 800

u x

99

ACTL3162 General Insurance Techniques Exercises

Plot of v against u, case theta= -0.987 Plot of y against x, case theta= -0.987

800

1.0

0.8

600

0.6

400

v

y

0.4

200

0.2

0.0

0

0.0 0.2 0.4 0.6 0.8 1.0 0 200 400 600 800

u x

800

1.0

0.8

600

0.6

400

v

y

0.4

200

0.2

0.0

0.0 0.2 0.4 0.6 0.8 1.0 0 200 400 600 800

u x

Plot of v against u, case theta= 0.962 Plot of y against x, case theta= 0.962

800

1.0

0.8

600

0.6

400

v

y

0.4

200

0.2

0.0

0.0 0.2 0.4 0.6 0.8 1.0 0 200 400 600 800

u x

100

ACTL3162 General Insurance Techniques Exercises

First, note:

the three different s considered for each copula correspond to negative dependence,

independence and positive dependence.

we are very close to the maximum negative and positive dependence possible for

FGM copulas: Spearmans rho needs to be within [1/3, 1/3] for such copulas.

Hence, the outputs give an idea of the whole spectrum of dependence FGM copulas

can offer: it is limited. . .

u v plots: in the independence cases, points are scattered in a random way and

no grouping is observed. For negative dependence, points are grouped more around

the top left and bottom right regions, whereas for positive dependence, points are

grouped more around the bottom left and top right regions, as they should be.

x y plots have been restricted to values up to 800the 96% quantile of the

Pareto(2, 200)in order to have a reasonable view of the observations. Otherwise,

displaying extreme data points results in aggregation of most of the data points

around the origin.

x y plots: if we had been presented with these plots only, it would have been im-

possible (or very difficult) to guess the structure of dependence between the random

variables, even after the modification described in the previous point. It is possible

to recognise some negative dependence when relevant (points are grouped close to

the axes and there are fewer points around the top right region) but the cases of

independence and positive dependence are not clear.

Overall it is more difficult to see the type of dependence in the x y scatterplots

than in the u v scatterplots, and this even though the marginal distributions are

the same. This shows that it is often more useful to plot probabilities rather than the

raw observations in order get some sense about the dependence structure between

random variables.

In both cases, positive or negative dependence are observed in both left and right

tails (there is no such thing as dependence only in one of the tails).

It seems that in the case of FGM copulas the maximum negative dependence is

weaker that the maximum positive dependence, whereas for Plackett copulas it

looks symmetrical.

If we needed a higher level of dependence than [1/3, 1/3], it would not be pos-

sible to use an FGM copula. On the other hand, the family of Plackett copulas is a

comprehensive family, that is, it can model the strongest negative and positive depen-

dence (countermonotonicity and comonotonicity copulas, respectivelythe Frechet

bounds). For instance, a Plackett can achieve a Spearmans rho of -0.8 and 0.8 if

the parameter is 0.0412 and 24.26, respectively. This is illustrated below:

101

ACTL3162 General Insurance Techniques Exercises

Plot of v against u, case theta= 0.0412 Plot of v against u, case theta= 24.26

1.0

1.0

0.8

0.8

0.6

0.6

v

v

0.4

0.4

0.2

0.2

0.0

0.0

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0

u u

102

Module 6

Tariffication

Exercise 6.1: [NLI13, Solution][?, Exercise 13] We would like to insure the following car fleet.

Assume that the car fleet can be modelled by a compound Poisson distribution (Vco means

(i) (i)

i vi i E[Y1 ] Vco(Y1 )

passager car 40 25% 2000 2.5

delivery van 30 23% 1700 2.0

truck car 10 19% 4000 3.0

coefficient of variation).

2. Calculate the premium for the car fleet using the variance loading principle with =

3 106 .

Exercise 6.2: [NLI13.1, Solution] Suppose that we have two independent lines of business, S1

and S2 , who have distributions S1 CompPoi(10, G1 Exp(10)) and S2 CompPoi(20, G2

Exp(20)).

(b) Calculate the premium using the variance loading principle: = E[S] + Var(S) using

= 0.25.

(c) Now assume

Pn that we have n lines of business (n being large), S1 , S2 , ..., Sn and the sum

S = i=1 Si . Propose a candidate for when using the standard deviation loading

principle: = E[S] + Var(S)1/2 (hint: Central Limit Theorem).

Exercise 6.3: [NLI7.3, Solution][?, Example 7.3] Denote Si,j as the claim of risk class (i, j) and

vi,j as the volume of business. Suppose we have the multiplicative tariff structure, i,j = 1,i 2,j

103

ACTL3162 General Insurance Techniques Exercises

for i = 1, 2, ..., I and j = 1, 2, .., J. We wish to find the parameters , 1,i and 2,j > 0 such

that the following expression is minimised,

(Si,j vi,j 1,i 2,j )2

X2 = . (6.1)

vi,j 1,i 2,j

We have a 2 by 4 table of risk classes. For simplicity, we can set vi,j 1 and we need to

determine the positive tariff factors , 1,i and 2,j for i = 1, 2, ..., I and j = 1, 2, .., J, that

minimise

(Si,j 1,i 2,j )2

X2 = . (6.2)

1,i 2,j

The observations of Si,j are given in the following table.

owned 1300 1200 1000 1200

leased 1800 1300 1300 1500

Exercise 6.4: [NLI21, Solution][?, Exercise 21] Provide design matrix Z for the pricing prob-

lem specified by the following risk class specification (assume a multiplicative tariff structure).

passager car 2000 1800 1500 1600

delivery van 2200 1600 1400 1400

truck 2500 2000 1700 1600

Exercise 6.5: [NLI21.1, Solution] Using the data from Table 6.1 (assuming a multiplicative

tariff structure). Further assume that wm = 1 and = w1. Also to ensure uniqueness of

solutions, we assume = 1 and 1,1 = 1. We follow a log-linear MLE structure for the claims.

(a) Provide the design matrix Z, the corresponding vector and log-claim vector X.

(b) Using the log-linear MLE formula (Equation (7.9) in ?), determine the tariffication factors.

6.4 Solutions

Solution 6.1: [NLI13, Exercise] Let S = 3i=1 Si be the aggregation of the three individual

P

lines of car fleet business, the moment generating function of S is

( 3 )

X

MS (r) = exp i vi MY (i) (r) 1 . (6.3)

1

i=1

1. Using the moment generating function of S, the expected claim amount of the car fleet is

3

X (i)

E[S] = i vi E[Y1 ] = 39330. (6.4)

i=1

104

ACTL3162 General Insurance Techniques Exercises

2. Using the moment generating function of S, the variance of the claim amount S is

3

X (i)

Var(S) = i vi E[(Y1 )2 ] = 693705000 (6.5)

i=1

The premium for the car fleet using the variance loading principle with = 3 106 is

(a) Using the aggregation of compound Poisson distributions, the sum S = S1 + S2 also has

a compound Poisson distribution

1 2

S = S1 + S2 CompPoi 30, G1 (x) + G2 (x) . (6.7)

3 3

1 1 2 1

E[S] = 30 + =2 (6.8)

3 10 3 20

1 2 2 2 3

Var[S] = 30 + = . (6.9)

3 102 3 202 10

Then the premium using the variance loading principle is = 2 + 0.25 0.3 = 2.075.

(c) According to the Central Limit Theorem, the sum S = ni=1 Si can be approximated by

P

the normal distribution with mean nE[Si ] = E[S] and variance nVar(Si ) = Var(S). A

natural candidate for is 1.96 (the 97.5% quantile of a standard normal random variable).

Such a premium corresponds to a 2.5% Value-at-Risk.

Solution 6.3: [NLI7.3, Exercise] Note that we need to set = 1 and 1,1 to ensure that we

have unique solutions.

R-code:

# first declare the matrix containing observations of S_ij

Sij<-c(1300,1800,1200,1300,1000,1300,1200,1500)

dim(Sij)<-c(2,4)

# define the objective function to be minimised

# note that we set mu and one of the weights to be 1

ObjFun<-function(x){

mu<-1

x1<-x[1]

x2<-x[2]

x3<-x[3]

x4<-x[4]

y1<-1

105

ACTL3162 General Insurance Techniques Exercises

y2<-x[5]

temp<-mu*t(c(x1,x2,x3,x4)%*%t(c(y1,y2)))

sum((Sij-temp)^2/temp)

}

# use optim function to product output

OptimResult<-optim(c(30,30,30,30,1),ObjFun,method = "L-BFGS-B",hessian=T)

OptimResult

# check eigenvalues and they are all positive

eigen(OptimResult$hessian)

# positive definite matrix -> local minimiser

1 1 0 0 1 0 0 0

2 1 0 0 0 1 0 0

3 1 0 0 0 0 1 0

4 1 0 0 0 0 0 1

5 0 1 0 1 0 0 0

6 0 1 0 0 1 0 0

7 0 1 0 0 0 1 0

8 0 1 0 0 0 0 1

9 0 0 1 1 0 0 0

10 0 0 1 0 1 0 0

11 0 0 1 0 0 1 0

12 0 0 1 0 0 0 1

There are multiply ways to convert the table above to be the design design matrix Z (as long

as it is full rank). If we choose to have an intercept term, then we need to drop one of the

three car-type columns and one of the four year-type columns (to ensure full-rankness). Here

we drop the columns truck and 51-60y. The resulting design matrix is

1 1 0 1 0 0

1 1 0 0 1 0

1 1 0 0 0 1

1 1 0 0 0 0

1 0 1 1 0 0

1 0 1 0 1 0

Z= . (6.10)

1 0 1 0 0 1

1 0 1 0 0 0

1 0 0 1 0 0

1 0 0 0 1 0

1 0 0 0 0 1

1 0 0 0 0 0

106

ACTL3162 General Insurance Techniques Exercises

(a) From Table 6.1, we can construct the following categorical variables

1 1 0 1 0 0 0 log(1300)

2 1 0 0 1 0 0 log(1200)

3 1 0 0 0 1 0 log(1000)

4 1 0 0 0 0 1 log(1200)

5 0 1 1 0 0 0 log(1800)

6 0 1 0 1 0 0 log(1300)

7 0 1 0 0 1 0 log(1300)

8 0 1 0 0 0 1 log(1500)

1,1 = 0. So the design matrix Z and have the following form

0 1 0 0 0

0 0 1 0 0

0 0 0 1 0 1,2

0 0 0 0 1 2,1

Z = 1 1 0 0 0 2,2 (6.11)

2,3

1 0 1 0 0

1 0 0 1 0 2,4

1 0 0 0 1

M LE = (Z 0 1 Z)1 Z 0 1 X, (6.13)

which yields

1,2

0.2227

2,1 7.2215

M LE = 2,2 = 7.0187

. (6.14)

6.9276

2,3

2,4 7.0903

Since we assume that 0 = 0 and 1,1 = 0, we now have a complete set of tariffication

factors,

= exp(0 ), 1,i = exp(1,i ) and 2,j = exp(2,j ). (6.15)

owned 1,1 = 1

leased 1,2 = 1.2495

2,1 = 1368 2,2 = 1117 2,3 = 1020 2,4 = 1200

107

Module 7

Exercise 7.1: [glm1, Solution] For the following members of the exponential dispersion family,

give the density (including the domain), the mean and the variance:

1. Normal(, 2 )

2. Poisson()

3. Binomial(m, p)

4. Negbin(r, p)

5. Gamma(, )

6. Inverse Gaussian(, )

Exercise 7.2: [glm2K, Solution] [?, Problem 8.6.8] The following is an extract from ?, pp.193

4.

density

ehy f (y)

fh (y) = R , (8.41)

ehz f (z)dz

provided the denominator is finite, i.e., the mgf with f (y) exists at h. A similar

transformation of the density can be performed for discrete distributions. In both

cases, the mgf with the transformed density equals mh (t) = m(t + h)/m(h). For a

density f in the exponential dispersion family, the cgf of fh has the form

b( + (t + h)) b() b( + h) b()

h (t) =

b( + h + t) b( + h)

= , (8.42)

which is again a cgf of an exponential dispersion family member with parameter

h = + h and the same .

It can be shown that the Esscher transform with parameter h < transforms

108

ACTL3162 General Insurance Techniques Exercises

2. Poisson(1) into Poisson(eh );

3. binomial(m,1/2) into binomial (m, eh (1 + eh )1 );

4. negative binomial (r, 1/2) into negative binomial (r, 1 eh /2) when < h <

log 2;

5. gamma (1,1) into (1,1 h) when < h < 1);

6. inverse Gaussian (1,1) into inverse Gaussian ( 1 2h,1 2h) when , h <

1/2.

So we see that all the examples of distributions in the exponential dispersion family

that we have given can be generated by starting with prototypical elements of each

type, and next taking Esscher transforms . . . .

Exercise 7.3: [glm3, Solution] [Jiwooks Final Exam Question 2002 - modified] The density

of the Binomial distribution is given by

n!

f (y; p) = py (1 p)(ny) .

(n y)!y!

Show that the Binomial distribution is a member of the exponential dispersion family with

density

y b()

f (y; , ) = exp + c(y; ) .

Exercise 7.4: [glm4K, Solution] [?, Problem 8.4.1] Verify that

D L(y; ) 2X yi

= 2 log = yi log (yi i )

L(y; y) i i

D L(y; ) 2X yi (yi i )

= 2 log = log +

L(y; y) i i i

Exercise 7.6: [NLI22, Solution][?, Exercise 22] Calculate the deviance statistics for the Poisson

and the gamma model, see also ?, equation (3.4).

109

ACTL3162 General Insurance Techniques Exercises

Exercise 7.7: [glm6K, Solution] [?, Problem 8.6.6] Show that in general, the scaled deviance

equals

D 2X

= wi {yi (i i ) [b(i ) b(i )]}.

i

Exercise 7.8: [glm7, Solution] Show that the deviance for an Inverse Gaussian distribution

has the following form:

n

X 1 (yi bi )2

D= 2 .

i=1

b i y i

Exercise 7.9: [glm8, Solution] Question #9, ACTL3003/5106 Final Exam 2005.

Exercise 7.10: [glm9, Solution] [Institute question, April 2006] An insurance company has a

set of n risks (i = 1, 2, ..., n) for which it has recorded the number of claims per month, Yij , for

m months (j = 1, 2, ..., m).

It is assumed that the number of claims for each risk, for each month, are independent Poisson

random variables with

E (Yij ) = ij .

These random variables are modelled using a Generalized Linear Model, with

n X m

X yij

2 yij log (yij y i )

i=1 j=1

y i

1

Pn

where y i = m j=1 yij .

3. A company has data for each month over a 2 year period. For one risk, the average

number of claims per month was 17.45. In the most recent month for this risk, there were

9 claims. Calculate the contribution that this observation makes to the deviance.

Exercise 7.11: [glm10, Solution] [Institute question, Sep 2003] There are m male drivers in

each of three age groups, and data on the number of claims made during the last year are

available. Assume that the numbers of claims are independent Poisson random variables. If

Yij is the number of claims for the jth male driver in group i (i = 1, 2, 3; j = 1, 2, ..., m), let

E(Yij ) = ij and suppose log (ij ) = i .

1. Show that this is a Generalized Linear Model, identifying the link function and the linear

predictor.

110

ACTL3162 General Insurance Techniques Exercises

3. For a particular data set with 20 observations in each group, several models are fitted,

with deviances as shown below:

Link function Deviance

, if i = 1, 2

Model 2 log (ij ) = 61.64

, if i = 3

whether or not model 1 is a significant improvement over model 2.

ii. Interpret these three models.

Exercise 7.12: [glm11, Solution] An insurance company tested for claim sizes under two

factors, i.e. CAR, the insurance group into which the car was placed, and AGE, the age

of the policyholder (i.e. two-way contingency table). It was assumed that the the claim size

yi follows a gamma distribution, i.e.

1 yi i i yi i

f (yi ) = exp for yi 0, i > 0, i = 1

(i ) yi i i

with a log-link function. Analysis of a set of data for which n = 8 provided the following SAS

output:

Observation Claim size CAR type Age group Pred Xbeta Resdev

1 27 1 1 25.53 3.24 0.30

2 16 1 2 24.78 3.21 1.90

3 36 1 1 3.41 1.03

4 45 1 2 38.09 3.64 1.11

5 38 2 1 40.85 3.71 0.46

6 27 2 2 36.97 3.61 1.73

7 14 2 1 2.45 0.69

8 6 2 2 14.59 2.68 2.55

111

ACTL3162 General Insurance Techniques Exercises

Exercise 7.13: [glm12R, Solution][R ] In this question, the vehicle insurance data set1 is used,

car.csv. This data set is based on one-year vehicle insurance policies taken out in 2004 or

2005. There are 67856 policies of which 4624 had at least one claim.

The data frame car.csv contains claim occurrence clm, which takes value 1 if there is a

claim and 0 otherwise. The variable veh value represents the vehicle value which takes value

from $0 $350, 000. We will not be concerned about other variables at the moment.

In this question, we will build a logistic regression model to apply to the vehicle insurance

data set. Previous study has shown that the relationship between the likelihood of occurrence

of a claim and vehicle value are possibly quadratic or cubic.

1. Suppose the relationship between vehicle value and the probability of a claim is cubic,

formulate the model and test significance of the coefficients.

2. Use AIC to determine the which model is the best model. Linear, quadratic or cubic.

Exercise 7.14: [glm13R, Solution][R ] Third party insurance 1 is a compulsory insurance for

vehicle owners in Australia. It insures vehicle owners against injury caused to other drivers,

passengers or pedestrians, as a result of an accident.

In this question, the third party claims data set Third party claims.xls is used. This data

set records the number of third party claims in a twelve-month period between 1984-1986 in

each of 176 geographical areas (local government areas) in New South Wales, Australia.

1. Now consider a model for the number of claims (claims) in an area as a function of the

number of accidents (accidents). Produce a scatter plot of of claims against accidents.

Do you think a simple linear regression model is appropriate?

2. Fit a simple linear regression to the model and use the plot command to produce residual

and diagnostic plots for the fitted model. What do the plots tell you?

3. Now fit a Poisson regression model with claims as response and log(accident) as the

predictor (include offset=log(population) in your code). Check if there is overdisper-

sion in the model by computing the estimate of .

estimates of the parameters and their standard errors.

7.4 Solutions

Solution 7.1: [glm1, Exercise] See the lecture notes (or Table 3.1 in ?). The density and the

mean are given in the table and the variance can be derived easily from the table with:

2 = V () .

Try to map some of the densities into the exponential family formulation.

1

?

112

ACTL3162 General Insurance Techniques Exercises

Solution 7.2: [glm2K, Exercise] (8.6.8) Although this has not been discussed in lecture, it

should not be a difficult exercise to show that exponential dispersion is preserved under

Esscher transformation. The proof is straightforward using the cgf argument in (8.42), although

mgf can also be used. Now, to prove the statements in Remark 8.6.10, start from

h (t) = et+h 1 eh 1 = eh et 1

which is clearly the cgf of a Poisson eh . As yet another example, in the Gamma(1, 1) case

from Table A, we have

1 1

h (t) = log log

1th 1h

1h

= log

1ht

which is clearly the cgf of a Gamma(1, 1 h) . For inverse Gaussian (1,1), we have

p

h (t) = (1 1 2(t + h)) (1 1 2h)

p

= 1 2h 1 2(t + h)

r

1 2t 2h

= 1 2h(1 )

1 2h

r

2t

= 1 2h(1 1 ).

1 2h

Solution 7.3: [glm3, Exercise] You ought to be able to verify that the Binomial belongs to

the family of Exponential Dispersion with

n!

b () = n log 1 + e , c (y; ) = log , and = 1.

(n y)!y!

p

= log = log .

1p n

The three components of a generalized linear model are: (1) Stochastic Component: The

observations Yi are independent and each follows an Exponential Dispersion P distribution. (2)

Systematic Component: Every observation has a linear predictor i = j xij j where xij

denotes the jth explanatory variable, and (3) Link function: The expected value E (Yi ) = i

is linked to the linear predictor i by the link function i = g (i ). Now to find the deviance of

113

ACTL3162 General Insurance Techniques Exercises

Q yi

n!

bi (nyi )

(ny )!y !

bi

n

(1 n

) yi nyi !

D i i i Y

bi n bi

= 2 log Q yi (nyi ) = 2 log

n! yi yi

(ny )!y ! n

(1 n ) i

yi n yi

i i

i

X

bi

n bi

= 2 yi log + (n yi ) log

i

yi n yi

n

X yi n yi

= 2 yi log + (n yi ) log .

i=1

b i n

b i

Solution 7.4: [glm4K, Exercise] (8.4.1) We know that if D denotes the deviance, the scaled

deviance is

D

= 2 log L/ b Le

by definition, where L

b is the likelihood computed using the MLEs b under the current model

replacing the , while L is the likelihood computed with the replaced by the estimates under

e

the full model, hence the actual observations y, in view of the remarks just below (8.22). To

show that (8.23) results from this is basic algebra. To see this, note that

n

Q bi yi

e bi /yi ! n yi !

D i=1

Y

bi

e(bi yi )

= 2 log n

= 2 log

yi

eyi yiyi /yi !

Q

i=1

i=1

n

X

bi

= 2 (b

i yi ) + yi log

i=1

yi

n

X

bi

= 2 i yi ) yi log

(b .

i=1

yi

Solution 7.5: [glm5K, Exercise] (8.4.2) To show that (8.26) results, following the discussion in

the previous problem, we can verify that, for exponential dispersion models, the scaled deviance

can be expressed as

D

= 2 log L/L b e

X yi i i

n e b b i b bi

e

= 2 .

i=1

n

D X yi (1/bi 1/yi ) log yi log

bi

= 2

i=1

n

2 X yi bi

= log (yi /b

i ) .

i=1

bi

Now, if the scale parameter were different for each observation according to some weight wi ,

then it is easy to verify.

114

ACTL3162 General Insurance Techniques Exercises

(Poisson) Let Xi be the number of claims of in a group of policies with Xi Poi(wi i ), where wi

represents the volume. Since we are more interested in the distribution of the claim frequency

Yi = Xi /wi whose probability mass function is

fYi (yi ; i ) = P(Xi = wi yi )

(wi i )wi yi

= ewi i

(wi yi )!

= exp [wi (yi log(wi i ) i ) log(wi yi )!] . (7.1)

So taking log to the above pmf, the deviance statistics for the Poison model is

X X

D(y, ) = 2 [wi (yi log(wi yi ) yi )] 2 [wi (yi log(wi i ) i )]

i i

X

=2 wi (yi log(yi /i ) yi i ). (7.2)

i

(w)w w1 wy

fY (y) = y e . (7.3)

(w)

We re-parametrise the distribution function via = / and = 1/. Then we have

w/

1 w

fY (y) = y w/1 ewy/()

(w/)

y/ log()

= exp + c(y, , w) . (7.4)

/w

Then the deviance statistics for the gamma model is

!

X yi /yi log(yi ) X y/i log(i )

D(y, ) = 2

i

/wi i

/wi

X

=2 wi (yi /i 1 log(yi /i )). (7.5)

i

Solution 7.7: [glm6K, Exercise] (8.6.6) Note that the log-likelihood can be expressed as

n

X yi i b (i )

` (; y) = + c (yi ; /wi )

i=1

/wi

D h

e y ` ;

i

= 2 ` ; by

X yi i b i

n e e yi i b bi

b

= 2 + c (yi ; /wi ) c (yi ; /wi )

i=1

/w i /w i

n

2 X h e i

= wi yi i b ei yi bi b bi

i=1

115

ACTL3162 General Insurance Techniques Exercises

Solution 7.8: [glm7, Exercise] Recall that the scaled deviance for any member of the Expo-

nential Dispersion family has the form

D h

e y ` ;

i

= 2 ` ; by

n

2 X h e i

= yi i b ei yi bi b bi

i=1

where for the Inverse Gaussian, we have verified (in lecture) that

1

= /2 , = / = , and b () = 2 = 1/.

22

Thus, the deviance can be expressed as

n

X 1 1 1 1

D = 2 yi 2 + yi 2

i=1

2y i y i 2b i

bi

n n

" 2 #

yi2

X

X 1 2yi 1 yi

= 2 1+ 2 = 1

i=1

2y i

b i

b i i=1

y i

bi

n

" 2 # n

X 1 bi yi X 1 1

= = 2

i y i ) 2 .

(b

i=1

yi

b i i=1

b y

i i

Solution 7.9: [glm8, Exercise] See Final Exams solution, Year 2005.

y

Q ijij eij

1. The likelihood is i,j and the log-likelihood is therefore

yij !

n X

X m

` (i ) = (yij log ij ij log yij !)

i=1 j=1

n X

X m

yij i ei log yij ! .

=

i=1 j=1

m

` (i ) X

= yij mei = 0

i j=1

so that m

i 1 X

e = yij = y i

m j=1

and the MLE is

bi = log y i .

116

ACTL3162 General Insurance Techniques Exercises

2. The deviance is

Pn Pm

i=1 j=1 (yij log yij yij log yij !)

2 [` (y; y) ` (y; )] = 2

ni=1 m

P P

j=1 (yij log y i y i log yij !)

n m

XX yij

= 2 yij log (yij y i ) .

i=1 j=1

yi

yij

Dij = 2 yij log (yij y i )

yi

9

= 2 9 log (9 17.45) = 4.98.

17.45

1. If Y has a Poisson distribution with mean parameter , then its density can be written

as

y y log

f (y; ) = e /y! = exp log y!

1

which is of the exponential dispersion family form. The link function is the log so that

g () = log and the linear predictor is

= log = i .

Y3 Ym yijij eij

i=1 j=1 yij !

so that the log-likelihood is

3 X

X m

(yij log ij ij log yij !) .

i=1 j=1

3

X 3

X

i

` (1 , 2 , 3 ) = me + yi+ i + constant

i=1 i=1

where yi+ refers to the sum of the observations in the ith group. Differentiating, we get

` (i )

= mei + yi+ = 0

i

so that the maximum likelihood estimator of i is

bi = log (yi+ /m) .

117

ACTL3162 General Insurance Techniques Exercises

3. In comparing the models, notice the nesting: Model 3 is the smallest and is contained in

Model 2 which is contained in Model 1. We may use our Rule of Thumb of significant

improvement if the decrease in deviance is larger than twice the additional parameter.

Here we summarize in table form:

First additional Significant

Model Deviance Difference d.f. D1 D2 > 2 (p q)? improvement?

Model 3 72.53 - -

Model 2 61.64 10.89 1 Yes Yes

Model 1 60.40 1.24 1 No No

improvement from Model 1.

Now, regarding interpretation of the models: Model 3 says that there is no difference in

the average number of claims for the three age groups. Model 2 says that there is no

difference in the average number of claims between age groups 1 and 2, but that the third

age group may be different. Model 1 gives the possibility of different average number of

claims for each age group.

Solution 7.12: [glm11, Exercise] We know that the linear predictor, for the ith observation,

is X

i = log i = xij j = xTi (in vector form).

j

Thus,

T

E (yi ) = i = exi .

and therefore, the predicted values are

and

E (y7 ) = e2.45 = 11.59.

ln = 0 + 1 x + 2 x2 + 3 x3

1

where x is the vehicle value and is the probability of a claim of the policy.

> car<-read.csv(".../car.csv")

> attach(car)

> names(car)

[1] "veh_value" "exposure" "clm" "numclaims" "claimcst0" "veh_body"

[7] "veh_age" "gender" "area" "agecat" "X_OBSTAT_"

> car.glm<-glm(clm~veh_value+I(veh_value^2)+I(veh_value^3),family=binomial,data=car)

> summary(car.glm)

118

ACTL3162 General Insurance Techniques Exercises

Call:

glm(formula = clm ~ veh_value + I(veh_value^2) + I(veh_value^3),

family = binomial, data = car)

Deviance Residuals:

Min 1Q Median 3Q Max

-0.4093 -0.3885 -0.3729 -0.3561 2.9462

Coefficients:

Estimate Std. Error z value Pr(>|z|)

(Intercept) -2.9247606 0.0476282 -61.408 < 2e-16 ***

veh_value 0.2605947 0.0420331 6.200 5.66e-10 ***

I(veh_value^2) -0.0382409 0.0084167 -4.543 5.53e-06 ***

I(veh_value^3) 0.0008803 0.0002752 3.199 0.00138 **

---

Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 . 0.1 1

Residual deviance: 33711 on 67852 degrees of freedom

AIC: 33719

The fits shows that all the coefficients are significant as the p-values are all smaller than

0.01.

2. > car.qua<-glm(clm~veh_value+I(veh_value^2),family=binomial,data=car)

> car.lin<-glm(clm~veh_value,family=binomial,data=car)

> car.lin$aic

[1] 33749.12

car.qua$aic

[1] 33718.92

car.cub$aic

[1] 33718.72

The difference between the AIC of the cubic and quadratic models is less than one. This shows

that if we include a cubic explanatory variable, the improvement of the fit quantified by AIC

only decreases by 0.2. Therefore, when evaluating a model by the principal of parsimony, a

quadratic model is preferred. Further, the AIC of the quadratic model is much less than that

of the linear, suggesting that the linear model is inadequate.

1. plot(accidents,claims,xlab="Accidents",ylab="Claims")

119

ACTL3162 General Insurance Techniques Exercises

We can clearly see that there is a concentration of points around the origin make it

difficult to discern the relationship between the predictor and response. The data is also

strongly heteroskedasitic, which means more variable for higher value of the predictor.

This is a violation of the homoskedasticity assumption of linear model.

2. > third.lm<-lm(claims~accidents,offset=log(population))

> plot(third.lm)

The residuals vs fitted plot shows that the residual is clearly do not follow a standard nor-

mal distribution and the variance seems to inflate as the fitted value increases. Diagnostic

checks indicate clear violation of the homoskedasticity assumption.

3. > third.poi<- glm(claims ~ log(accidents), family=poisson,offset=log(population))

> summary(third.poi)

Call:

glm(formula = claims ~ log(accidents), family = poisson, offset = log(population))

Deviance Residuals:

Min 1Q Median 3Q Max

-38.957 -3.551 0.116 3.842 45.965

Coefficients:

Estimate Std. Error z value Pr(>|z|)

(Intercept) -7.093809 0.026992 -262.81 <2e-16 ***

log(accidents) 0.259103 0.003376 76.75 <2e-16 ***

---

Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 . 0.1 1

Residual deviance: 15837 on 174 degrees of freedom

AIC: 17066

120

ACTL3162 General Insurance Techniques Exercises

> sum(resid(third.poi,type="pearson")^2)/third.poi$df.residual

[1] 101.7168

The estimate of takes a value of 101.7168. The inflated dispersion parameter suggests

there is overdispersion in the data.

4. > third.qpoi<- glm(claims ~ log(accidents), family=quasipoisson,offset=log(population))

> summary(third.qpoi)

Call:

glm(formula = claims ~ log(accidents), family = quasipoisson,

offset = log(population))

Deviance Residuals:

Min 1Q Median 3Q Max

-38.957 -3.551 0.116 3.842 45.965

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) -7.09381 0.27223 -26.058 < 2e-16 ***

log(accidents) 0.25910 0.03405 7.609 1.66e-12 ***

---

Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 . 0.1 1

Residual deviance: 15837 on 174 degrees of freedom

AIC: NA

Poisson =1 -7.09381(0.02699) 0.25910(0.003376)

Quasi-Poisson =101.7172 -7.09381(0.27223) 0.25910(0.03405)

The quasi-poisson estimates of are identical to those of the Poisson model, but with

standard errors larger by a factor of 1/2 = 10.085.

121

Module 8

Theory

Let Xjt denote the claim size of policy j during year t, for 1 j J and 1 t T . This

random variable is function of a risk profile, which cannot be observed and which is assumed to

be the same for a given contract (for given j but across t) but different between policies (across

j).

The unobservable risk profile is modelled as a random variable , and the risk profile of policy

j is a possible outcome j of (but we cannot observe it). Risk profiles across contracts are

assumed to be independent. We will denote

as the expectation and variance of claim sizes, as functions the risk profile, respectively. The

moments of these quantities are key quantities and are denoted

Note that when we will need to consider () or 2 () for a particular policy j, we will write

j () or j2 (). Finally, assume that for the same policy (given a certain risk profile ), claim

sizes are independent over time (across t). Claim sizes across policies (across j) are always

independent.

For nonparametric estimates, we will denote

T J

1X 1X

Xj = Xjt and X = Xj

T t=1 J j=1

as the average claim size of policy j and overall average claim size (across j as well), respectively.

Finally Xj,T +1 is the claim size whose expectation we want to estimate for policy j for the T + 1

period.

8.1 Preliminaries

Exercise 8.1: [cre1, Solution] Prove that

122

ACTL3162 General Insurance Techniques Exercises

and that

Cov (X, Y ) = Cov (X, Y ) .

Also, derive the formula

Cov(X, Y ) = E [Cov(X, Y |Z)] + Cov (E[X|Z], E[Y |Z])

for the decomposition into conditional variances.

a

if i = j, t 6= k,

Cov (Xit , Xjk ) = a + s2 if i = j, t = k,

0 if i 6= j.

s2

Cov(Xjt , X j ) = a + ,

T

and then find Cov(Xj,T +1 , X j ).

(

s2

a+ T

if i = j,

Cov(X i , X j ) =

0 6 j.

if i =

s2

1

Cov(X j , X) = a+ .

J T

s2

1

V ar(X) = a+ .

J T

Exercise 8.7: [cre6R, Solution][R ] Suppose you are given the following observed claims for 3

groups and 6 years:

t = 1 t = 2 t = 3 t = 4 t = 5 t = 6 Xj

j = 1 1047 1874 1501 1497 1876 1740 1589

j = 2 2003 1726 1524 1776 1764 2010 1800

j = 3 1597 943 920 1780 980 1010 1205

Test whether the mean claims per group are all equal. What does this imply regarding premium

calculations?

Exercise 8.8: [cre3K, Solution] [?, Problem 7.5.2] Suppose that has a gamma(, ) prior

distribution, and that given = , the annual numbers of claims X1 , . . . , XT are indepen-

dent Poisson() random variables. Prove that the posterior distribution of , given X1 =

x1 , . . . , XT = xt , is gamma( + x , + T ), where x = x1 + . . . + xT .

123

ACTL3162 General Insurance Techniques Exercises

Exercise 8.9: [cre4, Solution] [April 2006 Institute of Actuaries CT6 Question] An insurer

has for 2 years insured a number of domestic animals against veterinary costs. In year 1, there

were n1 policies and in year 2, there were n2 policies. The number of claims per policy per year

follows a Poisson distribution with unknown (mean) parameter .

Individual claim amounts were a constant c in year 1 and a constant c(1 + r) in year 2. The

average total claim amount per policy was y1 in year 1 and y2 in year 2. Prior beliefs about

follow a Gamma distribution with mean / and variance /2 . In year 3, there are n3

policies, and individual claim amounts are c(1 + r)2 . Let Y3 be the random variable denoting

the average total claim amounts per policy in year 3.

1. State the distribution of the number of claims on the whole portfolio over the 2 year

period.

2. Derive the posterior distribution of , given y1 and y2 .

3. Show that the posterior expectation of Y3 given y1 and y2 can be written in the form of

a credibility estimate

Z k + (1 Z) c (1 + r)2

specifying expressions for k and Z.

4. Describe k in words and comment on the impact the values of n1 , n2 have on Z.

Exercise 8.10: [cre5, Solution] You are given that an individual automobile insured has an

annual claim frequency distribution that follows a Poisson distribution with mean , where

because of parameter uncertainty, actually follows a Gamma distribution with parameters

and . A total of one claim is observed for the insured over a five-year period.

One actuary assumes that = 2 and = 5, and a second actuary assumes the same

mean for the Gamma distribution, but only half the variance.

Both actuaries determine the Bayesian premium for the expected number of claims in the

next year using their model assumptions.

Determine the ratio of the Bayesian premium that the first actuary calculates to the

Bayesian premium that the second actuary calculates.

Exercise 8.11: [cre7, Solution] Assume that a company only has one policy j and the claim

sizes Xj1 , Xj2 , ..., XjT are identically distributed, and conditionally on , Xj1 , Xj2 , ..., XjT are

independent and identically distributed with

E [Xjt |] = () and V ar (Xjt |) = s2 ().

We are interested in a linear estimator for Xj,T +1 of the form

P = g0 + gX j .

Find values for g0 and g such that P is unbiased and such that hit minimisesi the quadratic

2

error with respect to Xj,T +1 . You may assume that minimising E Xj,T +1 P is equivalent

h i2

to minimising E () P .

124

ACTL3162 General Insurance Techniques Exercises

Exercise 8.12: [cre8, Solution] In the Buhlmann model, find the variance of the credibility

premium

P = zX j + (1 z)X

as well as its MSE (remember we want to estimate Xj,T +1 ).

Exercise 8.13: [cre9, Solution] In the Buhlmann model, recall that M SB: mean square be-

tween is

J

T X 2

M SB = Xj X

J 1 j=1

and M SW : mean square within is

J T

1 XX 2

M SW = Xjt X j .

J(T 1) j=1 t=1

E[M SB] = aT + s2 and that E[M SW ] = s2 .

Cov XT +1 , X

z= .

V ar X

can be re-expressed as

T

z= .

T +k

Give the expression for the constant k and explain how it will affect the credibility coefficient

z.

[This is a past exam question]

Exercise 8.15: [cre11K, Solution] [?, Problem 7.4.1] Let X1 , . . . , XT be independent random

2

variables with variances

P 2 V2 ar(Xt ) = s /wt for certain positive

P numbers wt , t = 1, . . . , T . Show

that the variance t t s /wt of the linear combination t t Xt with = 1 is minimal when

we take t wt , where the symbol means proportional to. Hence the optimal solution has

t = wt /w . Prove also that the minimal value of the variance in this case is s2 /w .

Exercise 8.16: [cre12K, Solution] [?, Problem 7.4.2] Prove that in the Buhlmann-Straub

model, we have V ar(X zw ) V ar(X ww ). (Here,

T

X zj

X zw = Xjt ,

z

t=1

T

X wjt

X jw = Xjt

t=1

w j

J

X wj

X ww = Xjw ,

j=1

w

and zj is the credibility factor.)

125

ACTL3162 General Insurance Techniques Exercises

Exercise 8.17: [cre13K, Solution] [?, Problem 7.4.10] Estimate the credibility premiums in

the Buhlmann-Straub setting when the claims experience for three years is given for three

contracts, each with weight wjt 1. The claims on the contracts are as follows:

j=1 10 12 14

j=2 13 17 15

j=3 14 10 6

1 2 3 4 5

risk class 1 v1,t 729 786 872 951 1019

S1,t 583 1100 262 837 1630

X1,t 80.0% 139.9% 30.0% 88.0% 160.0%

risk class 2 v2,t 1631 1802 2090 2300 2368

S2,t 99 1298 326 463 895

X2,t 6.1% 72.0% 15.6% 20.1% 37.8%

risk class 3 v3,t 796 827 874 917 944

S3,t 1433 496 699 1742 1038

X3,t 180.0% 60.0% 80.0% 190.0% 110.0%

risk class 4 v4,t 3152 3454 3715 3859 4198

S4,t 1765 4145 3121 4129 3358

X4,t 56.0% 120.0% 84.0% 107.0% 80.0%

risk class 5 v5,t 400 420 422 424 440

S5,t 40 0 169 1018 44

X5,t 10.0% 0.0% 40.0% 240.1% 10.0%

Table 8.1: Observed claim Si,t and corresponding numbers of policies vi,t .

\

\

(a) Choose the data of Table 8.1 and calculate the inhomogeneous credibility estimator ( i)

for the claims ratios under the assumption that the collective mean is given by 0 = 90%

and the variance between risk classes is given by 2 = 0.20.

(b) What changes if the variance between risk classes is given by 2 = 0.05?

Exercise 8.19: [NLI24, Solution][?, Exercise 24] Estimate the prediction uncertainty E[(Xi,T +1

hom

\

\

(i ) )2 ] for the data in Table 8.1 under the assumption that the volume grows 5% in each

risk class.

The observed numbers of policies vi and claim counts Ni in 21 different regions are given in Table

8.2. Calculate the inhomogeneous credibility estimators for each region i under the assumption

that Ni |i has a Poisson distribution with mean (i )vi = i 0 vi and E[i ] = 1. The prior

frequency parameter is given by 0 = 8.8% and the prior uncertainty by 2 = 2.4 104 .

126

ACTL3162 General Insurance Techniques Exercises

region i vi Ni

1 50,061 3,880

2 10,135 794

3 121,310 8,941

4 35,045 3,448

5 19,720 1,672

6 39,092 5,186

7 4,192 314

8 19,635 1,934

9 21,618 2,285

10 34,332 2,689

11 11,105 661

12 56,590 4,878

13 13,551 1,205

14 19,139 1,646

15 10,242 850

16 28,137 2,229

17 33,846 3,389

18 61,573 5,937

19 17,067 1,530

20 8,263 671

21 148,872 15,014

total 763,525 69,153

Table 8.2: Observed volumes vi and claims counts Ni in regions i = 1, 2, ..., 21.

8.4 Solutions

Solution 8.1: [cre1, Exercise]

We first have

= E (XY ) E (X) E (Y ) + E (XZ) E (X) E (Z)

= Cov (X, Y ) + Cov (X, Z)

Remember E[X + Y ] = E[X] + E[Y ] holds irrespective of the dependence structure of X and

Y . Then we have

= E (XY ) E (X) E (Y )

= Cov (X, Y ) .

= E (XY ) E [E (X |Z ) E (Y |Z )]

= E (XY ) E (X) E (Y ) + E (X) E (Y ) E [E (X |Z ) E (Y |Z )]

= Cov (X, Y ) {E [E (X |Z ) E (Y |Z )] E [E (X |Z )] E [E (Y |Z )]}

= Cov (X, Y ) Cov [E (X |Z ) , E (Y |Z )] .

127

ACTL3162 General Insurance Techniques Exercises

Notice that when X = Y , then it reduces to the familiar formula for conditional variances:

or

V ar (X) = E [V ar (X |Z )] + V ar [E (X |Z )] .

Using the result of Exercise 8.1, we have

Cov(Xit , Xjk ) = E [Cov(Xit , Xjk |)] + Cov [E(Xit |), E(Xjk |)]

= E [Cov(Xit , Xjk |)] + Cov [i (), j ()] .

= E [V ar(Xit |)] + V ar [E(Xit |)]

= E s2 () + V ar [()]

= s2 + a.

For the last case, since policies are independent across multiple lines (when i 6= j), we have

Cov(Xit , Xjk ) = 0.

Firstly, we have

Cov(Xjt , X j ) = E Cov Xjt , X j | + Cov E [Xjt |] , E X j |

1

= E [Cov (Xjt , Xjt |)] + Cov (E [Xjt |] , E [Xjt |])

T !

T

1 X 1

since Cov Xjt , X j | = Cov Xjt , Xit = Cov Xjt , Xjt

T T

" # t=1

T

1X

since E X j | = E Xjt = E[Xjt |]

T t=1

1

= E [V ar(Xjt |)] + Cov (j (), j ())

T

= s2 /T + a.

Cov(Xj,T +1 , X j ) = E Cov Xj,T +1 , X j | + Cov E [Xj,T +1 |] , E X j |

= 0 + Cov (E [Xj,T +1 |] , E [Xjt |])

= Cov (j (), j ()) = a.

128

ACTL3162 General Insurance Techniques Exercises

For the case of i = j, we have

!

1X

Cov X j , X j = Cov Xjt , X j

T all t

1X

= Cov(Xjt , X j )

T all t

1

= T (s2 /T + a) = s2 /T + a.

T

For the case of i 6= j, since risk profiles across contracts are assumed to be independent,

therefore Cov(X i , X j ) = 0.

We have

Cov(X j , X) = E Cov X j , X| + Cov E X j | , E X|

" !# " #!

1 X 1 X

= E Cov X j , Xj + Cov E X j | , E Xj

J

all j

J

all j

1 1

= E V ar(X j |) + Cov E X j | , E X j |

J " J !#

t=T

1 1 X 1

= E V ar Xjt + Cov (E[Xjt |], E[Xjt |])

J T t=1 J

11 1

= E [V ar (Xjt |)] + V ar (E[Xjt |])

JT J

1 s2

= +a .

J T

We have

V ar X = Cov(X, X)

( )

1 X X

= Cov(X i , X j )

J 2 all i all j

2

1 s

= J +a (because of the result in Exercise 8.4)

J2 T

1 s2

= +a .

J T

Xjt = m + j + jt = mj + jt

129

ACTL3162 General Insurance Techniques Exercises

for j = 1, ..., J and t = 1, ..., T , where the error terms jt N (0, s2 ), we can test equality of

means with

H0 : m1 = m2 = m3

or equivalently

H0 : 1 = 2 = 3 = 0.

Here, J = 3 and T = 6. The test statistic is

1

P 2

M SB J1

T Xj X

j

F = = 2 .

M SW 1

P P

X jt X j

J(T 1) j t

> summary(aov.cred1)

Df Sum Sq Mean Sq F value Pr(>F)

factor(j) 2 1093732 546866 5.9099 0.01280 *

Residuals 15 1388006 92534

---

Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 . 0.1 1

The p-value = 1.280% which is less than a level of significance of 5% suggests evidence to

reject the null hypothesis. Alternatively, we may compare this with the F -value from a table:

F0.05 (2, 15) = 3.68. The observed F -statistics of 5.9099 is larger, therefore we reject the null.

The groups are therefore non-homogeneous. This suggests that we should be asking for a

different premium for each group. A premium credibility formula which attaches credibility

factor according to each own groups claim experience would be suitable.

= c fX1 | (x1 |) fXT | (xT |) f ()

T

Y e xi 1 1

= c e

i=1

xi ! ()

T x

= c e 1 e

= c e(T + ) (x +)1

where c is a constant that makes this a proper density. Clearly this has the form of a Gamma

density with parameters x + and T + .

Solution 8.9: [cre4, Exercise] Exact solutions drawn from the Institute paper/report.

1. The total number of claims has a Poisson distribution with parameter (n1 + n2 ) .

130

ACTL3162 General Insurance Techniques Exercises

2. Let Yi denote the average total claim amount per policy in year i and let Xi denote the

total number of claims in year i. Then Xi has a Poisson distribution with parameter ni

and

n1 n2

X1 = Y1 and X2 = Y2 .

c c (1 + r)

We have

f ( |y1 , y2 ) f (y1 , y2 | ) ()

en1 (n1 )y1 n1 /c en2 (n2 )y2 n2 /c(1+r) e 1

e(n1 +n2 +) (+y1 n1 /c+y2 n2 /c(1+r))1

which implies that the posterior distribution of is Gamma with parameters + y1 n1 /c +

y2 n2 /c(1 + r) and n1 + n2 + .

3. Thus, our predicted value of Y3 , given the observed claims y1 and y2 is given by

c (1 + r)2

E (Y3 |y1 , y2 ) = E (X3 |y1 , y2 )

n3

c (1 + r)2 + y1 n1 /c + y2 n2 /c(1 + r)

= n3

n3 n1 + n2 +

2 2

c (1 + r) + n1 y1 (1 + r) + n2 y2 (1 + r)

=

n1 + n2 +

= c (1 + r)2

n1 + n2 +

!

n1 y1 (1 + r)2 + n2 y2 (1 + r) n1 + n2

+

n1 + n2 n1 + n2 +

n1 y1 (1 + r)2 + n2 y2 (1 + r)

k=

n1 + n2

and

n1 + n2

Z= .

n1 + n2 +

4. k is effectively a weighted average of the inflation adjusted average claim amounts for the

previous 2 years, weighted by the number of policies in force. As the number of policies

in force increases, Z becomes closer to 1, and so the more weight is placed on the actual

experience and less on the prior expectations.

Solution 8.10: [cre5, Exercise] It was shown in class that when claim frequency X1 , ..., XT are

independent Poisson() with having a Gamma(, ) prior, then the posterior distribution is

Gamma( + x , + T ) so that the Bayesian premium is given by

+ x

E ( |X1 , ..., XT ) = .

+T

According to the first actuary, = 2 and = 5. The second actuary sets the parameter with

equal mean, but only half the variance. Therefore

2 1

= and 2 =

5

( ) 25

131

ACTL3162 General Insurance Techniques Exercises

so that = 4 and = 10. Since there is only one claim in 5 years, x = 1 and T = 5. The

first actuary sets the premium to

+ x 2+1 3

= =

+T 5+5 10

and the second actuary to

+ x 4+1 5 1

= = = .

+ T 10 + 5 15 3

The ratio is therefore

3/10

= 90%

1/3

which interestingly, despite assuming larger variance, the first actuary has a smaller premium.

Intuitively, this is because as the increasing number of years contributes to larger credibility

attached to ones own claims experience, it also provides a greater number of opportunities

to correct for premium miscalculations in the past. Presumably, at policy inception, the first

actuary will require larger premium than the second actuary, in the absence of any claims

experience. If claims experience becomes more favorable than expected, then there will be

larger correction in premium calculated. And the magnitude of correction increases then with

time, assuming of course, favorable experience continues.

The MSE of the estimator P can be expressed as

h i2

M SE = E Xj,T +1 P

2

= E Xj,T +1 g0 + gX j

2

= V ar Xj,T +1 g0 + gX j + E Xj,T +1 g0 + gX j .

Unbiasedness condition implies the second term is zero. Now, note that

V ar Xj,T +1 g0 + gX j = V ar (Xj,T +1 ) 2Cov Xj,T +1 , g0 + gX j + V ar g0 + gX j

= V ar (Xj,T +1 ) 2gCov Xj,T +1 , X j + g 2 V ar X j .

2Cov Xj,T +1 , X j + 2gV ar X j = 0

Cov Xj,T +1 , X j

g=

V ar X j

g0 + gE X j = E (Xj,T +1 ) = E [ ()]

or equivalently

g0 = (1 g) E [ ()] .

132

ACTL3162 General Insurance Techniques Exercises

!

Cov Xj,T +1 , X j Cov Xj,T +1 , X j

P = Xj + 1 E [ ()]

V ar X j V ar X j

Cov Xj,T +1 , X j

z= .

V ar X j

Using Exercise 8.3 and 8.4 yields the familiar credibility formula:

Cov Xj,T +1 , X j V ar [ ()]

z = = 1

V ar X j T

E [ 2 ()] + V ar [ ()]

T V ar [ ()] T

= 2

= 2

.

E [ ()] + T V ar [ ()] T + (E [ ()] /V ar [ ()])

Solution 8.12: [cre8, Exercise] From Exercise 8.5 and 8.6, we obtained V ar(X) and Cov(X j , X).

Now, we have the variance of P

V ar zX j + (1 z) X

= z 2 V ar X j + 2z (1 z) Cov X j , X + (1 z)2 V ar X

a + s2 /T

= z 2 (a + s2 /T ) + 2z(1 z)(a + s2 /T ) + (1 z)2

J

= (a + s2 /T ) z 2 + 2z(1 z) + (1 z)2 /J

The expression in curly braces is inferior to 1 as long as J > 1, which shows that the variance

of the credibility premium is lower than the one of X j . Furthermore, choosing z = 0 or z = 1

yields the variances of X and X j , respectively, as it should.

The MSE (as an estimator for Xj,T +1 ) of the credibility premium can be derived as follows:

h 2 i

E Xj,T +1 zX j (1 z) X

= V ar Xj,T +1 zX j (1 z) X

[because unbiasedness of the linear estimator]

= V ar (Xj,T +1 ) 2Cov Xj,T +1 , zX j + (1 z) X + V ar zX j + (1 z) X

V ar (Xj,T +1 ) = s2 + a.

Cov Xj,T +1 , zX j + (1 z) X = zCov Xj,T +1 , X j + (1 z)Cov Xj,T +1 , X

1z

= zCov Xj,T +1 , X j + Cov Xj,T +1 , X j

J

a

= za + (1 z)

J

133

ACTL3162 General Insurance Techniques Exercises

s2 (1 z)2

2 a 2

M SE = s + a 2za 2(1 z) + (a + ) z + 2z(1 z) +

J T J

2

2(1 z) (1 z)

= a 1 2z + z 2 + 2z(1 z) +

J J

2

1 2 (1 z)

+s2 1 + z + 2z(1 z) +

T J

2

2z z 2 (1 z)2

2 1z 2

= a 1z +s 1+ +

J T JT

2 2

(1 z)2

(1 z )(J 1) 2 2z z

= a +s 1+ + .

J T JT

J

T X h 2 i

E (M SB) = E Xj X .

J 1 j=1

J

T X h 2 i

E (M SB) = E Xj X

J 1 j=1

J

T X h 2 i 2

= E Xj X E[X j X]

J 1 j=1

J

T X

= V ar X j X

J 1 j=1

J

T X

= V ar X j 2Cov(X j , X) + V ar X

J 1 j=1

J

s2 s2 s2

T X 1 1

= a+ 2 a+ + a+

J 1 j=1 T J T J T

J

s2

T X 1

= 1 a+

J 1 j=1 J T

s2

T

= (J 1) a + = aT + s2

J 1 T

J T

1 XX h 2 i

E (M SW ) = E Xjt X j

J (T 1) j=1 t=1

134

ACTL3162 General Insurance Techniques Exercises

J T

1 XX h 2 i

E (M SW ) = E Xjt X j

J (T 1) j=1 t=1

J T

1 XX h 2 i 2

= E Xjt X j E[Xjt X j ]

J (T 1) j=1 t=1

J T

1 XX

= V ar Xjt X j

J (T 1) j=1 t=1

J T

1 XX

= V ar (Xjt ) + V ar X j 2Cov Xjt , X j

J (T 1) j=1 t=1

J T

1 XX

(a + s2 ) + a + s2 /T 2 a + s2 /T

=

J (T 1) j=1 t=1

J T

1 XX

= s2 (1 1/T )

J (T 1) j=1 t=1

1

= JT s2 (1 1/T ) = s2 ,

J (T 1)

which proves the desired result.

Cov Xj,T +1 , X j

z= .

V ar X j

s2

Cov Xj,T +1 , X j = a and V ar X j = a + .

T

Thus, we have the familiar credibility formula:

Cov Xj,T +1 , X j a T

z= = 2 = s2

V ar X j a + sT T+ a

and k is therefore

E [s2 ()] s2

k= = .

V ar [ ()] a

Recall the formula of the credibility premium,

PTcred

+1 = zX j + (1 z)m.

Firstly notice that, if we have more experience in the data(i.e. T increases), then z will increase.

This makes sense as more experience means that we will give more credibility to the individual

mean X j .

If the heterogeneity of the portfolio increases (a %) that is risks are quite different amongst

135

ACTL3162 General Insurance Techniques Exercises

each portfolio, then we will expect k to decrease. Decreased value of k will increase the credi-

bility coefficient z. So if the portfolios we have are quite different from each other, we will use

more information from the individual mean structure X j , i.e. giving it more credibility.

In another situation where if the risk variability decreases within the portfolio(s2 &), then

we will also expect k to decrease which will result in a increasing value of z. This again makes

sense. If each individual portfolio does not vary dramatically, then we will obviously use more

information on the mean structure of individual portfolio X j .

!

X X

V ar t Xt = t2 V ar (Xt )

t t

where V ar (Xt ) = s2 /wt and subject to the condition = t t = 1. The Lagrangian for

P

this problem can be written as

X

t2 s2 /wt ( 1) .

L=

t

L

= 2t s2 /wt = 0

t

and is true if and only if

t

= 2.

wt 2s

Since this equality must hold for every t, t must then be proportional to wt . This implies

jt = wjt /wj . The variance is therefore

T

! T

X X

(wjt /wj )2 s2 /wjt = s2 /wj .

V ar jt Xjt =

t=1 t=1

Solution 8.16: [cre12K, Exercise] In Exercise 8.15 we show that if the variance is of the

form s2 /wt , then the alphas of an estimator of X should be proportional to the inverse: wt

(or equivalently, wt /s2 since s2 is a constant). We have for the unconditional variance of Xij

(whose expectation is the collective premium, which we need in the credibility premium)

which is now proportional to zj and not to wjt . The best unconditional expected value of Xij

should then be computed using the zj s, not the wij s, as this will minimise the variance of the

estimator.

wjt wj

Note that X jw = Tt=1 wj Xjt andXww = Jj=1 w

P P

Xjw .

136

ACTL3162 General Insurance Techniques Exercises

Consider the sequence of independent random variables X1w , , XJw . The variance of X jw is

T T

X wjt X wjt

= V ar(E[ Xjt |]) + E[V ar( Xjt |)]

t=1

wj t=1

wj

T T 2

X wjt X wjt

= V ar( E[Xjt |]) + E[ V ar(Xjt |)]

t=1

wj t=1

wj

T T 2 2

X wjt X wjt ())

= V ar( ()) + E[ ]

t=1

w j t=1

w j w j t

T

X wjt

= V ar(()) + 2

E[ 2 ()]

t=1

w j

a

= a + s2 /wj = ,

zj

z

Using the result in Exercise 8.15 above, we can see that when j = zj , the variance of the

linear combination Jj=1 j Xjw is minimized.

P

z

Hence the variance of the linear combination X zw = Jj=1 zj Xjw is the smallest among all the

P

linear combinations. Therefore,

J

X wj

V ar(X zw ) V ar( Xjw ) = V ar(X ww ).

j=1

w

PT wjt PJ wj

Solution 8.17: [cre13K, Exercise] Note that X jw = t=1 wj Xjt and X ww = j=1 w Xjw .

From the formulas we saw in the lecture we get

1 X

se2 = wjt (Xjt Xjw )2 = 8

J (T 1) j,t

and

wj (Xjw Xww )2 (J 1) se2

P

j 11

a= P 2 = .

w j wj /w

e

3

Thus, the Buhlmann-Straub credibility factor is given by

aT

e 11

ze = 2

= .

aT + se

e 19

The credibility premiums are therefore:

X

3

b2,T +1 = ze 15 + (1 ze) 12 1 = 13.88

X

3

b3,T +1 = ze 10 + (1 ze) 12 1 = 10.98.

X

3

137

ACTL3162 General Insurance Techniques Exercises

Solution 8.18: [NLI23, Exercise] From ?, Theorem 8.17, the inhomogeneous credibility esti-

mator is given by

\

\

( i ) = i,T Xi,1:T + (1 i,T )0 , (8.1)

b

with credibility weight i,T and observation based estimator Xbi,1:T . Since 2 and 0 are given,

2 2

we only need to obtain the estimator for , T = 261.2. Then using the above formula, results

are summarised in the following table,

risk class 1 risk class 2 risk class 3 risk class 4 risk class 5

bi,T 76.94% 88.64% 76.94% 93.37% 61.72%

\

\ 2

( i ) ( = 0.2) 0.9866506 0.3702171 1.1623246 0.8988722 0.7169975

\

\ 2

(i ) ( = 0.05) 0.9512165 0.5048693 1.0550632 0.8990594 0.8148146

R-code:

# need MASS package for calculating column and row sums

library(MASS)

# read the data first

v.matrix<-c(729,1631,796,3152,400,786,1802,827,3454,420,872,

2090,874,3715,422,951,2300,917,3859,424,1019,2368,944,4198,440)

dim(v.matrix)<-c(5,5)

S.matrix<-c(583,99,1433,1765,40,1100,1298,496,4145,0,262,326,

699,3121,169,837,463,1742,4129,1018,1630,895,1038,3358,44)

dim(S.matrix)<-c(5,5)

X.matrix<-S.matrix/v.matrix

w.matrix<-v.matrix

tau2<-0.2

mu0<-0.9

Xhat<-rep(rowSums(w.matrix*X.matrix)/rowSums(w.matrix),5)

dim(Xhat)<-c(5,5)

s2hat<-rowSums(w.matrix*((X.matrix-Xhat)^2))/4

sigma2hat<-mean(s2hat)

# calculating \hat{alpha}_{i,T}

alphahat<-rowSums(w.matrix)/(rowSums(w.matrix)+sigma2hat/tau2)

# calculating \hat{X}_{i,1:T}

Xhat<-rowSums(w.matrix*X.matrix)/rowSums(w.matrix)

alphahat*Xhat+(1-alphahat)*mu0

tau2<-0.05

mu0<-0.9

138

ACTL3162 General Insurance Techniques Exercises

Xhat<-rep(rowSums(w.matrix*X.matrix)/rowSums(w.matrix),5)

dim(Xhat)<-c(5,5)

s2hat<-rowSums(w.matrix*((X.matrix-Xhat)^2))/4

sigma2hat<-mean(s2hat)

# calculating \hat{alpha}_{i,T}

alphahat<-rowSums(w.matrix)/(rowSums(w.matrix)+sigma2hat/tau2)

# calculating \hat{X}_{i,1:T}

Xhat<-rowSums(w.matrix*X.matrix)/rowSums(w.matrix)

alphahat*Xhat+(1-alphahat)*mu0

Solution 8.19: [NLI24, Exercise] From ?, Equation (8.18), the prediction uncertainty has the

following formula is given by,

!

hom 2

\ 2

\ 1 i,T

E Xi,T +1 ( i)

= 2

+ (1 i,T ) 1 + P . (8.2)

wi,T +1 i i,T

We have estimators T2 = 261.2, T2 = 0.1021 and wi,T +1 using 5% increment factor. Results

are tabulated below,

risk class 1 risk class 2 risk class 3 risk class 4 risk class 5

hom

\

\

E[(Xi,T +1 (i) )2 ] 0.2482469 0.1062646 0.2676411 0.0597071 0.5744316

R-code:

# need MASS package for calculating column and row sums

library(MASS)

# read the data first

v.matrix<-c(729,1631,796,3152,400,786,1802,827,3454,420,872,2090,

874,3715,422,951,2300,917,3859,424,1019,2368,944,4198,440)

dim(v.matrix)<-c(5,5)

S.matrix<-c(583,99,1433,1765,40,1100,1298,496,4145,0,262,326,

699,3121,169,837,463,1742,4129,1018,1630,895,1038,3358,44)

dim(S.matrix)<-c(5,5)

X.matrix<-S.matrix/v.matrix

w.matrix<-v.matrix

sigma2hat<-261.2

tau2hat<-0.1021

# calculating \hat{alpha}_{i,T}

alphahat<-rowSums(w.matrix)/(rowSums(w.matrix)+sigma2hat/tau2hat)

139

ACTL3162 General Insurance Techniques Exercises

w.matrix[,5]*1.05

sigma2hat/(w.matrix[,5]*1.05)+(1-alphahat)*tau2hat*(1-alphahat)/sum(alphahat)

In this exercise, we have 21 risk classes and for every risk class we have T = 1 observation.

The data is summarised in Table 8.2 where claims counts are denoted as Ni and the observed

numbers of policies are vi . We consider a BS model on the frequencies Xi = Ni /vi where the

conditional expectation and variance is given by (using the Poisson assumption),

E[Ni |i ]

(i ) = E[Xi |i ] = = i 0 , (8.3)

vi

2 (i ) Var[Ni |i ] i 0

= Var[Xi |i ] = 2

= . (8.4)

vi vi vi

The collective mean is 0 = E[(1 )] = 0 E[1 ] = 8.8%. The prior uncertainty is given by

2 = 2.4104 . The volatility within risk classes also happens to be 2 = E[ 2 (1 )] = 0 E[1 ] =

8.8% (thanks to the Poisson assumption).

1 50061 3880 0.07750544 0.9927289 0.07758175

2 10135 794 0.07834238 0.9650849 0.07867957

3 121310 8941 0.07370373 0.9969865 0.07374682

4 35045 3448 0.09838779 0.9896456 0.09828023

5 19720 1672 0.08478702 0.9817458 0.08484567

6 39092 5186 0.13266141 0.9907076 0.13224640

7 4192 314 0.07490458 0.9195671 0.07595788

8 19635 1934 0.09849758 0.9816682 0.09830514

9 21618 2285 0.10569895 0.9833217 0.10540377

10 34332 2689 0.07832343 0.9894328 0.07842568

11 11105 661 0.05952274 0.9680372 0.06043295

12 56590 4878 0.08619898 0.9935624 0.08621057

13 13551 1205 0.08892333 0.9736546 0.08889900

14 19139 1646 0.08600240 0.9812020 0.08603995

15 10242 850 0.08299160 0.9654371 0.08316471

16 28137 2229 0.07921953 0.9871362 0.07933248

17 33846 3389 0.10013000 0.9892827 0.10000000

18 61573 5937 0.09642213 0.9940803 0.09637228

19 17067 1530 0.08964669 0.9789679 0.08961205

20 8263 671 0.08120537 0.9575109 0.08149407

21 148872 15014 0.10085174 0.9975431 0.10082016

R-code:

# read the data first

v.vector<-c(50061,10135,121310,35045,19720,39092,4192,19635,21618,

34332,11105,56590,13551,19139,10242,28137,33846,61573,17067,8263,148872)

140

ACTL3162 General Insurance Techniques Exercises

N.vector<-c(3880,794,8941,3448,1672,5186,314,1934,2285,2689,661,

4878,1205,1646,850,2229,3389,5937,1530,671,15014)

x.vector<-N.vector/v.vector

# collective mean, vol between, vol within

mu0<-0.088

sigma2<-0.088

tau2<-0.00024

# credibility weights

alpha.vector<-v.vector/(v.vector+sigma2/tau2)

# credibility estimators

ans<-cbind(v.vector,N.vector,x.vector,

alpha.vector,alpha.vector*x.vector+(1-alpha.vector)*mu0)

141

Module 9

Claims Reserving

Exercise 9.1: [IBNR1K, Solution] [?, Problem 9.2.1] Consider the following simple situation:

0 1 2 3 4

1 A1 A2 A3 B1 E

2 A4 A5 A6 B2

3 C1 C2 C3 X

b34

4 D1 D2 D X

b44

5 F

D B

X44 =

A

(where the sum D includes D ) and

D (B + X34 )

X44 =

A + C

indeed produces the same estimate.

Exercise 9.2: [IBNR1Y, Solution] Suppose that an insurer has observations of claims from

year 2009 to year 2011. The observations of the claims arrivals and development are recorded in

two tables: the first one lists the accident date, reporting date, settlement status and settlement

date (if applicable) of each observed claim; the second one records the transaction of the claims.

1 2009.01.31 2009.02.01 settled 2010.05.02

2 2009.07.12 2009.12.11 unsettled NA

3 2010.02.15 2010.05.19 settled 2010.06.20

4 2010.12.08 2011.02.08 settled 2011.08.13

5 2011.04.05 2011.06.20 unsettled NA

6 2011.08.22 2011.09.11 settled 2011.12.31

142

ACTL3162 General Insurance Techniques Exercises

1 2009.03.15 100

1 2010.01.25 50

2 2010.02.20 60

2 2011.05.10 180

3 2010.05.21 160

3 2010.06.13 70

4 2011.03.22 80

4 2010.08.10 40

5 2011.09.06 90

6 2011.10.12 500

6 2011.11.09 200

Construct a 3-by-3 annual loss triangle based on the above two tables.

Exercise 9.3: [IBNR3K, Solution] [?, Problem 9.2.3] Apply the chain ladder method to the

given IBNR triangle with cumulated figures. What could be the reason why run-off triangles

to be processed through the chain ladder method are usually given in a cumulated form?

Origin 0 1 2 3 4

1 232 338 373 389 391

2 258 373 429 456

3 221 303 307

4 359 430

5 349

Exercise 9.4: [IBNR4K, Solution] [?, Problem 9.2.4] Apply the arithmetic separation method

to the same data of the previous exercise. Determine the missing values by linear or loglinear

interpolation, whichever seems more appropriate.

Exercise 9.5: [IBNR5, Solution] For a certain portfolio of general insurance policies, denote by

Xij the claims that occur in accident year i, but paid in development year j, where i = 1, 2, ..., t

and j = 0, 1, ..., t 1 with observable claims only for i + j t. The triangle below shows the

observed incremental claims for this portfolio for a 3-year development period:

Accident Development Year

Year 0 1 2

1 2,541 1,029 217

2 2,824 790

3 1,981

1. Give five (5) reasons for the possible delay between the occurrence and the actual payment

of claims that gives rise to Incurred-but-not-Reported (IBNR) reserves.

2. In the Chain Ladder approach of estimating the bottom-half of the claims run-off triangle,

the claims Xij are assumed to be Poisson distributed with mean i j . Derive explicit

forms for the maximum likelihood estimators for the parameters i and j .

143

ACTL3162 General Insurance Techniques Exercises

3. Using the result in (2), calculate the maximum likelihood estimates for i and j for

i = 1, 2, 3 and j = 0, 1, 2 and use these to estimate the bottom half of the triangle.

4. Explain the difference between the Chain Ladder approach and the Arithmetic Separation

Method.

Exercise 9.6: [IBNR7, Solution] Estimate the expected outstanding claims reserve for the

data in the table below (figures in $1000), using the Bornhuetter-Ferguson method. Assume

that an expected loss ratio of 85%, and that the total claims paid are $1, 942, 000.

Accident Earned

Year premium 0 1 2 3

2003 860 473 620 690 715

2004 940 512 660 750

2005 980 611 700

2006 1,020 647

Exercise 9.7: [IBNR6, Solution] [Jiwooks Final Exam Question 2002] The claim payments

(in incremental form) from a general insurance portfolio are represented by the following table

Development Year

2000 2001 2002

Year of Origin

2000 X00 X01 X02

2001 X10 X11

2002 X20

Based on the following assumptions, we would like to estimate the outstanding claims where:

claim payments for each year of origin and development year have a log-normal distribu-

tion,

i+j j

!

X X

ln (Xij ) Normal + k + k , 2

k=1 k=1

where Xij denotes the claim amount paid in development year j arising from losses occuring

in year of origin i.

claim payments for each year of origin and development year are independent.

the expected value of the logarithm of the claim payments in year of origin 0, Year 2000

and development year 0, Year 2000 is .

the expected change in the logarithm of the claim payments from one accounting year to

the next is given by i for each accounting year i.

144

ACTL3162 General Insurance Techniques Exercises

for each year of origin, the expected change in the logarithm of the claims payment from

development year j 1(j = 1, 2...) to development j is equal to j and this is the same

for each year of origin.

the logarithm of claim payments have the same variance, 2 , regardless of year of origin

or year of development.

The i values allow for any inflation in values from one accounting year to the next. The j

values allow for the settlement pattern of claims over time arising from the same policy year.

The run-off triangle of expected values for the logarithm of the claims payments will then be

Development Year

2000 2001 2002

Year of Origin

+1

2000 +1 +2

+1 +1

+2

+1

2001

+1 +2

+1

2002 +1

+2

Assuming that 3 = 4 = 0.018, estimate the outstanding claims X12 , X21 and X22 .

Exercise 9.8: [IBNR2Y, Solution] Consider a 3-by-3 incremental loss triangle where we have

I = t = 3 and J = 2 (that is, we observe {xij ; i + j 3, 1 i 3, 0 j 2}, and all the

observations are not cumulative). The exposures of the ith accident period is a known constant

ci (1 i 3). We assume that Xij follows a normal distribution with parameters ij = j ci

and ij = j (1 i 3 and 0 j 2). In other words, the probability density function is of

Xij is

(xij j ci )2

1

2 2

fXij (xij ) = e j . (9.1)

j 2

mean and variance of the outstanding claims liability.

145

ACTL3162 General Insurance Techniques Exercises

Exercise 9.9: [IBNR3Y, Solution] Use the ChainLadder package in R and the USAApaid

data, perform the following tasks:

2. calculate the conditional mean square error of prediction (MSEP) of the reserving esti-

mates with Macks formula;

3. calculate the process uncertainty and parameter uncertainty involved in the above condi-

tional MSEP;

5. estimate the reserves with a Poisson model and compared the results to the Chain-Ladder

reserves; comment on the comparison.

The USAApaid data required for this exercise comes from the private passenger auto liabil-

ity/medical line of business of the United Services Automobile Association company (https:

//cran.r-project.org/web/packages/ChainLadder/ChainLadder.pdf). This is part of the

Schedule P dataset maintained by the National Association of Insurance Commissioners. The

Schedule P dataset provides real insurance data of nice lines of business over 10 years of all

U.S. general insurers. One can refer to the Casualty Actuarial Society website via http:

//www.casact.org/research/index.cfm?fa=loss_reserves_data for more information and

a clean subset of the data.

9.5 Solutions

Solution 9.1: [IBNR1K, Exercise] (9.2.1) This is immediate from, beginning with equation,

b44 = D (B + ) = D (B + C B /A )

X

A + C A + C

D B /A (A + C )

=

A + C

D B

=

A

which gives the result.

Solution 9.2: [IBNR1Y, Exercise] The construction and result of the loss triangle are shown

in the following table.

146

ACTL3162 General Insurance Techniques Exercises

of Origin 0 1 2

1 100 (50+60=)110 180

2 (160+70+40)=270 80

3 (90+500+200)=790

Solution 9.3: [IBNR3K, Exercise] (9.2.3) First, it can be verified that the row and column

totals are:

Year Development Year Row

of Origin 0 1 2 5 4 Total

1 232 338 373 389 391 391

2 258 373 429 456 456

3 221 303 307 307

4 359 430 430

5 349 349

Column Total 1419 374 95 43 2

You can proceed by estimating the parameters as suggested in the book (or applying the

mechanics of using ratios of cumulative claims as discussed in lecture). We have

b1 = 391.0,

b2 = 458.3,

b3 = 325.1,

b4 = 498.0, and

b5 = 545.5

and

b0 = 0.640, b1 = 0.224, b2 = 0.081, b3 = 0.051, and b4 = 0.0051.

As a result, we have the bottom part of the claims run-off triangle:

of Origin 0 1 2 3 4 Total

1

2 2.3 2.3

3 16.5 1.7 18.2

4 40.3 25.2 2.5 68.0

5 122.0 44.1 27.6 2.8 196.5

Some differences may exist due to rounding. The total IBNR reserve is about 285.

bk in the Arithmetic Separation

method are:

and

b1 = 346.2170,

b2 = 413.0731,

b3 = 390.0336,

b4 = 515.2749, and

b5 = 453.

These can be obtained as suggested by the maximum likelihood estimates derived in lecture.

For example, P

i+j=5 xij

X

b5 = P4 = xij = 349 + 71 + 4 + 27 + 2 = 453

j=0 j i+j=5

147

ACTL3162 General Insurance Techniques Exercises

the sum of the claims in the big diagonal, and so on. You must be able to verify the rest. Then,

we can extrapolate the -factors linearly to yield k , for k = 6, ..., 9:

b6 = 518.27,

b7 = 549.85,

b8 = 581.43,

b9 = 613.01 k = 328.79 + 31.58 k)

(b

One can also exponentially extrapolate, leading to

bk = e5.81+0.0759k .

b6 = 526.05,

b7 = 567.53,

b8 = 612.29,

b9 = 660.57

(The value X32 = 4 arouses suspicion.) As a result, the lower right triangle of estimated values

for the first case (linear extrapolation) becomes:

Year Development Year Row

of Origin 0 1 2 3 4 Total

1

2 2.3 2.3

3 23.0 2.4 25.4

4 36.3 24.4 2.6 63.3

5 109.4 38.5 25.8 2.7 176.4

The total IBNR reserve required would be about 267.4. (Try working out the exponential case!)

Alternatively, one can use linear extrapolation based on the last two -factors:

k5

bk = 5

b5 + (b b4 ) = 5

b5 + (b b4 )(k 5); k = 6, ..., 9.

54

One can also use log-linear extrapolation based on the last two -factors:

k5

ln

bk = ln b5 ln

b5 + (ln b4 ) = ln b5 ln

b5 + (ln b4 )(k 5); k = 6, ..., 9.

54

1. Some possible reasons for delay: (1) delay in assessing exact size or amount of claims; (2)

delay in investigating whether claim is valid; (3) long legal proceedings; (4) claims have

occurred, but not filed later; and (5) claim consists of series of payments (e.g. disability

insurance).

2. First, notice that you can write the probability:

P (X = xij ) = ei j (i j )xij /xij !

for i, j satisfying i + j < t. The full likelihood of all observed values can be written as

Y

L (, ) = ei j (i j )xij /xij !

i,j

Take the log of the likelihood and maximize. The solutions will have the form:

P

j xij

P

xij

bi = P and bj = Pi ,

bj j i

bi

P the denominators here are actually

row sums (Ri = j xij ) and column sums (Kj = i xij ), respectively.

148

ACTL3162 General Insurance Techniques Exercises

3. Notice that we can actually write the observed claims in the chain ladder approach as

Year 0 1 2

1 1 0 1 1 1 2

2 2 0 2 1

3 3 0

So easily we can verify, together with the assumption that all claims settle after 3 devel-

opment years, i.e. 0 + 1 + 2 = 1, the following:

217

2 = = 0.0573

3787

2824 + 790

2 = = 3, 834

1 0.0573

1029 + 790

1 = = 0.2387

3787 + 3834

1981

3 = = 2, 814

1 0.0573 0.2387

2541 + 2824 + 1981

0 = = 0.7040

3787 + 3834 + 2814

Thus using these results to estimate the bottom half, we would have:

X b2 b2 = 220; X

b22 = b3 b1 = 672; X

b31 = b3 b2 = 161.

b22 =

4. The chain ladder method assumes that the claims are Poisson distributed, Xij Poisson(i j ),

where the s denote the accident year effect, and the s.denote the development year

effect. It has no calendar year effect, unlike the Arithmetic Separation method where

the claims are assumed to be Poisson(j k ). Here as in the chain ladder, s.denote

the development year but the s denote the accident year effect. Both methods use

maximum likelihood to estimate their corresponding parameters, although in predicting

unpaid claims, because future calendar years have not occurred yet, in the Arithmetic

separation method, the s may have to be extrapolated from the estimated ones.

Solution 9.6: [IBNR7, Exercise] First calculate the initial expected total loss as 85% of the

earned premium. This gives figure of 731, 799, 833 and 867.

Now calculate the development factors for individual years in the usual way. We find that the

factors are 1.2406, 1.250, 1.0362.

Tackling the years one at a time:

The total expected outgo for Accident Year 2005 is 715 as we are assuming that Accident Year

2005 is fully run-off.

For Accident Year 2006, the expected outgo was initially 799. On this basis we would expect

799

to have paid out 1.0362 = 771.09 so far. So we should have to pay out 799 771.09 = 27.91 in

the future. In fact we have incurred 750, so our final figure would be 750 + 27.91 = 777.91.

For Accident Year 2007, the expected outgo was initially 833. On this basis we would expect to

833

have paid out 1.03621.125 = 714.58 so far. So we would have to pay out 833 714.58 = 118.42

in the future. In fact we have incurred 700 so far, so our finial figure should be 700 + 118.42 =

818.42.

149

ACTL3162 General Insurance Techniques Exercises

For Accident Year 2008, the expected outgo was initially 867. On this basis we would expect

867

to have paid out 1.03621.1251.2406 = 599.50 so far. So we would have to pay out 867 599.50 =

267.5 in the future. In fact we have incurred 647 so far, so our final figure would be 647+267.5 =

915.5.

So the total payout expected is 3225.83, of which we have already paid 1942. So the balance is

1284.

Solution 9.7: [IBNR6, Exercise] (Jiwooks solution to Final Exam Question 2002) Note that

we can write the claims run-off model as:

Development Year

0 1 2

Year of Origin

+1

0 +1 +2

+1 +1

+2

+1

+1 +2

1

+1 +2 +3

+1 +1

+2

+1

+1 +2

2 +1 +2 +3

+2 +3 +4

+1 +1

+2

Firstly,

12 = + 1 + 2 + 3 + 1 + 2

= 7.0632 + 0.025 + 0.012 + 0.018 0.028 + 0.0123

= 7.1097

Hence,

log X12 N ormal 7.1097, 12

So 2

b12 = E (X12 ) = e+ 2

X =7.1097+0.5

= e7.6097 = 2017.67

Secondly,

21 = + 1 + 2 + 3 + 1

= 7.0632 + 0.025 + 0.012 + 0.018 0.0208

= 7.0974

Hence

log X21 N ormal 7.0974, 12

150

ACTL3162 General Insurance Techniques Exercises

So 2

b21 = E (X21 ) = e+ 2 = e7.5974 = 1993.01

X

Lastly,

22 = + 1 + 2 + 3 + 4 + 1 + 2

= 7.0632 + 0.025 + 0.012 + 0.018 + 0.018 0.0208 + 0.0123

= 7.1277

So 2

b22 = E (X22 ) = e+ 2 = e7.6277 = 2054.32

X

3j

2 Y (xij j ci )2

Y 1

2 2

L(0 , 1 , 2 , 0 , 1 , 2 ) = e j . (9.2)

j=0 i=1

j 2

3j

2 X

(xij j ci )2

X

log L(0 , 1 , 2 , 0 , 1 , 2 ) = ln j

j=0 i=1

2j2

2 2 P3j

(9.3)

2

i=1 (xij j ci )

X X

= ((3 j) ln j )

j=0 j=0

2j2

gives

P3j P3j

log L(0 , 1 , 2 , ) i=1 (xij ci ) j i=1 c2i

= . (9.4)

j j2

log L(0 ,1 ,2 ,)

Solving j

= 0 gives the maximum likelihood estimate of j , denoted by j ,

P3j

i=1 (xij ci )

j = P 3j 2 . (9.5)

i=1 ci

(2)

151

ACTL3162 General Insurance Techniques Exercises

Solution 9.9: [IBNR3Y, Exercise] Please see below the R codes with results.

install.packages(ChainLadder)

require(ChainLadder)

data(USAApaid)

ResultMCL<- MackChainLadder(USAApaid,est.sigma="Mack")

summary(ResultMCL)

# > summary(ResultMCL)

# $ByOrigin

# Latest Dev.To.Date Ultimate IBNR Mack.S.E CV(IBNR)

# 0 886334 1.0000000 886334.0 0.0000 0.000 NaN

# 1 982148 0.9992023 982932.1 784.0531 916.992 1.1695534

# 2 1075537 0.9966965 1079101.8 3564.7739 1327.393 0.3723639

# 3 1138375 0.9927425 1146697.2 8322.1732 1666.579 0.2002577

# 4 1226650 0.9843548 1246146.2 19496.1724 2945.469 0.1510794

# 5 1324732 0.9633905 1375072.8 50340.7522 7486.013 0.1487068

# 6 1320130 0.9164206 1440528.5 120398.4954 19876.901 0.1650926

# 7 1185300 0.8284440 1430754.6 245454.5883 34104.399 0.1389438

# 8 966162 0.6636149 1455907.7 489745.7092 63485.591 0.1296297

# 9 542021 0.3454994 1568804.4 1026783.4153 117999.863 0.1149219

#

# $Totals

# Totals

# Latest: 1.064739e+07

# Dev: 8.442082e-01

# Ultimate: 1.261228e+07

# IBNR: 1.964890e+06

# Mack S.E.: 1.491160e+05

# CV(IBNR): 7.589026e-02

ResultMCL$Mack.ProcessRisk[,ncol(USAApaid)]

# > ResultMCL$Mack.ProcessRisk[,ncol(USAApaid)]

# 0 1 2 3 4 5

# 0.0000 631.4351 969.2396 1269.0693 2450.1816 6582.9597

# 6 7 8 9

# 17940.2609 31237.4293 58835.1132 109898.2894

ResultMCL$Total.ProcessRisk[ncol(USAApaid)]

152

ACTL3162 General Insurance Techniques Exercises

# > ResultMCL$Total.ProcessRisk[ncol(USAApaid)]

# 10

# 129958.2

ResultMCL$Mack.ParameterRisk[,ncol(USAApaid)]

# > ResultMCL$Mack.ParameterRisk[,ncol(USAApaid)]

# 0 1 2 3 4 5

# 0.0000 664.9542 906.9437 1080.2545 1634.7471 3564.4124

# 6 7 8 9

# 8557.9334 13686.9669 23850.5710 42968.9854

ResultMCL$Total.ParameterRisk[ncol(USAApaid)]

# > ResultMCL$Total.ParameterRisk[ncol(USAApaid)]

# [1] 73119.55

# 1-year CDR

CDR(ResultMCL)

# > CDR(ResultMCL)

# IBNR CDR(1)S.E. Mack.S.E.

# 0 0.0000 0.000 0.000

# 1 784.0531 916.992 916.992

# 2 3564.7739 1035.163 1327.393

# 3 8322.1732 1134.252 1666.579

# 4 19496.1724 2478.155 2945.469

# 5 50340.7522 6874.055 7486.013

# 6 120398.4954 18413.859 19876.901

# 7 245454.5883 28014.262 34104.399

# 8 489745.7092 53622.597 63485.591

# 9 1026783.4153 98121.714 117999.863

# Total 1964890.1331 125036.244 149116.032

ResultPoisson <- glmReserve(USAApaid)

summary(ResultPoisson)

# > summary(ResultPoisson)

# Latest Dev.To.Date Ultimate IBNR S.E CV

# 1 982148 0.9992024 982932 784 2330.911 2.97310136

# 2 1075537 0.9966963 1079102 3565 4509.067 1.26481552

# 3 1138375 0.9927426 1146697 8322 6562.979 0.78863004

# 4 1226650 0.9843550 1246146 19496 9699.278 0.49750094

153

ACTL3162 General Insurance Techniques Exercises

# 6 1320130 0.9164209 1440528 120398 23501.424 0.19519779

# 7 1185300 0.8284437 1430755 245455 34460.994 0.14039638

# 8 966162 0.6636147 1455908 489746 53506.953 0.10925450

# 9 542021 0.3454995 1568804 1026783 106222.199 0.10345146

# total 9761055 0.8324323 11725945 1964890 142114.888 0.07232715

One can see that the reserves estimates obtained from the Poisson model and the CL model

are the same. One can prove this by setting = 1 in the over-dispersed Poisson model and

follow the same steps as those in section 9.3.2.

The joint likelihood function of observations Dt and parameters is

J1

Y (fj (j 1))j j 1 j fj (j 1)

h(Dt , ) = j e

j=0

(j )

g(C1,0 , . . . , CI,0 )

(9.8)

Ci,j

j 2 Ci,j

J1 (tj1)I j

1 j C

Y Y j2 2 2 i,j+1

Ci,j+1 j

e j .

Ci,j

j=0 i=1 2

j

(fj (j 1))j 1 j fj (j 1)

(j )

j j e calculates the likelihood of j = j ; there are J of them, with

j = 0, . . . , J;

g(C1,0 , . . . , CI,0 ) represents the likelihood of observing the first column, which we dont

necessarily need to know;

! Ci,j

j 2 Ci,j

j

2

1 j C

j 2 2 i,j+1

j

! Ci,j+1 e j calculates the likelihood of Ci,j+1 given Ci,j . The range of j is

Ci,j

2

j

from 0 to J 1 since the first column is not included here. Given the value of j, the range

of i is from 1 to min(t j 1, I).

Ci,j

J1 (tj1)I

2 j2 Ci,j+1

jj 1 ej fj (j 1)

Y Y

j

h(|Dt ) h(Dt , ) j e j

j=0 i=1

! (9.9)

P(tj1)I Ci,j Ci,j+1

J1

Y j + i=1 1 j fj (j 1)+ (tj1)I

P

2 i=1 2

j

= j e j

.

j=0

independent Gamma random variables. In particular, the posterior density function is

(tj1)I (tj1)I

X Ci,j X Ci,j+1

fj |Dt Ga j + 2

, fj (j 1) + 2

. (9.10)

i=1

j i=1

j

154

ACTL3162 General Insurance Techniques Exercises

(t)

Ci,J =E[Ci,J |Dt ]

=E[E[Ci,J |Ci,J1 ]|Dt ]

=E[Ci,J1 1

J1 |Dt ]

..

.

J1 (9.11)

Y

=E[Ci,ti 1

j |Dt ]

j=ti

J1

Y

=Ci,ti fjt

j=ti

j |Dt ].

1. A higher business volume tends to result in lower uncertainty (measured by both total

msep1/2 and CDR msep1/2 ) as a percentage of reserves. This can be explained by higher

diversification of claims associated with a higher business volume.

2. The ratio of the one-year claims development result uncertainy over the total run-off

uncertainty stays relatively stable across varying companies. This is because the ratio

depends on the nature of the business and here we are concerned with the same business

segment of different companies. This shows that knowing the next diagonal releases a

major part of the claims run-off risks, that is, around 80% of the total run-off uncertainty.

(This is similar to Exercise 9.18)

155

Module 10

Exercise 10.1: [DnG1, Solution] [Decisions & Games notes, exercise # 2] The loss function

under a decision problem is summarized below:

states

decision

1 2 3

d1 14 12 13

d2 13 15 14

d3 11 15 5

2. Given the probability distribution: P (1 ) = 41 , P (2 ) = 14 , and P (3 ) = 21 , determine the

Bayes criterion solution.

Exercise 10.2: [DnG2, Solution] [Decisions & Games notes, exercise # 3] The profit per

client-day made by a privately owned health center depends on the variable costs involved.

Variable costs, over which the owner of the health center has no control, take one of three

levels: 1 =high, 2 =most likely, and 3 =low. The owner has to decide at what level to set

the number of client-days for the coming year. Client-days can be either d1 = 16, d2 = 13.4, or

d3 = 10 (each in 000). The profit in ($) per client-day is as follows:

states

decision

1 2 3

d1 85 95 110

d2 105 115 130

d3 125 135 150

1. Determine the Bayes criterion solution based on the annual profits, given the probability

distribution P (1 ) = 0.1, P (2 ) = 0.6, and P (3 ) = 0.3.

2. Determine both the minimax solution and the maximin solution to this problem.

Exercise 10.3: [DnG4, Solution] [Decisions & Games notes, exercise # 5] A firm is contem-

plating three investment alternatives: stocks, bonds, and a savings account, involving three

156

ACTL3162 General Insurance Techniques Exercises

potential economic conditions: accelerated, normal, or slow growth. Each condition has a

probability of occurrence P (accelerated growth)= 0.2, P (normal growth)= 0.5 and P (slow

growth)= 0.3. It is assumed that the decision maker who has $100,000 to invest wishes to

invest all the fund in a single investment class. The annual returns ($) yielded from the stocks,

bonds, and savings account are as follows:

Economic Conditions

Alternative accelerated normal slow

Investment growth growth growth

Stocks 20,000 13,000 -8,000

Bonds 16,000 12,000 2,000

Savings 10,000 10,000 10,000

2. Determine both the minimax regret solution and maximin solution to this problem.

3. Explain briefly when Bayesian decision analysis (i.e. Bayes rule) can be used.

Exercise 10.4: [DnG3, Solution] [Decisions & Games notes, exercise # 4] Consider the fol-

lowing two-person, zero-sum game:

Player B

strategies

x y z

Player A 1 250 300 150

strategies 2 50 165 125

3 100 275 225

1. Using the rule of dominance, reduce the payoff matrix to a 2-by-2 matrix.

2. Solve algebraically for the mixed-strategy probabilities for players A and B and determine

the expected gain for player A and the expected loss for player B. Discuss the meaning

of this solution value.

10.3 Solutions

Solution 10.1: [DnG1, Exercise] Decisions & Games, Exercise # 2: Solution from Institute.

First, check out the table below:

states

decision p1 = 0.25 p2 = 0.25 p3 = 0.5 expected

1 2 3 maximum loss

d1 14 12 13 14 13

d2 13 15 14 15 14

d3 11 15 5 15 9

157

ACTL3162 General Insurance Techniques Exercises

Thus, the minimax solution is d1 . The Bayes criterion solution is d3 since it gives smallest

expected loss. [Refer to bold values.]

Solution 10.2: [DnG2, Exercise] Decisions & Games, Exercise # 3: Solution from Institute.

First convert the table into annual profits as follows:

annual profits

decision p1 = 0.1 p2 = 0.6 p3 = 0.3 expected

1 2 3 maximum regret minimum profit

d1 1360 1520 1760 47 1360 1576

d2 1407 1541 1742 18 1407 1587.9

d3 1250 1350 1500 260 1250 1385

Therefore, Bayes criterion solution is to choose d2 as it gives the largest expected profit. The

minimax is to choose d3 and the maximin solution is to choose d2 .

Solution 10.3: [DnG4, Exercise] Decisions & Games, Exercise # 5: Jiwooks solution, Emil

modified.

Economic Conditions

probabilities 0.2 0.5 0.3

Alternative accelerated normal slow Expected

Investment growth growth growth profit

Stocks 20,000 13,000 8,000 8,100

Bonds 16,000 12,000 2,000 9,800

Savings 10,000 10,000 10,000 10,000

where for example, 8,100 = 20,000 0.2 + 13,000 0.5 + (8,000) 0.3. Bayes Decision:

choose Savings.

2. The opportunity loss (amount of regret) are given below:

Economic Conditions

Alternative accelerated normal slow maximum

Investment growth growth growth regret

Stocks 0 0 18,000 18,000

Bonds 4,000 1,000 8,000 8,000

Savings 10,000 3,000 0 10,000

Thus, for minimax Decision: choose Bonds (i.e. choose minimum of maximum regret).

Economic Conditions

Alternative accelerated normal slow minimum

Investment growth growth growth payoff

Stocks 20,000 13,000 8,000 8,000

Bonds 16,000 12,000 2,000 2,000

Savings 10,000 10,000 10,000 10,000

So for maximin Decision: choose Savings.

158

ACTL3162 General Insurance Techniques Exercises

3. Bayes rule can be used when we want to revise probabilities of potential states of nature

based on additional information, experiments and personal judgments.

Solution 10.4: [DnG3, Exercise] Decisions & Games, Exercise # 4: Jiwooks solutions.

Player B

x y z

250 < 300 > 150

1

50 < 165 > 125

Player A 2

3 100 < 275 > 225

So Strategy 1 & 3 dominate 2 for Player A. Strategy x & z dominate y for Player B.

Therefore, a 2-by-2 matrix would be

Player B

x z

Player A 1 250 150

3 100 225

2. If Player B selects strategy x, the possible payoffs for Player A are 250 and 100. Therefore,

if Player A selects strategy 1 with probability p, Player As expected gain is given by

250p + 100(1 p)

150p + 225(1 p)

So, we have

225p = 125

125 5 4

p = = ,1 p =

225 9 9

Similarly, for Player Bs expected gain, we need

225p 75

3 6

p = ,1 p =

9 9

Hence, Player As gain will be:

5 4

if B selects x : 250 + 100 = 183.33

9 9

5 4

if B selects z : 150 + 225 = 183.33

9 9

159

ACTL3162 General Insurance Techniques Exercises

3 6

if A selects1 : 250 + 150 = 183.33

9 9

3 6

if A selects 3 : 100 + 225 = 183.33

9 9

No matter what strategy player B chooses, if player A employs the mixed strategy that is

composed with ( 59 1st , 49 3rd ), it is guaranteed for him to obtain 183.33 (in expected value

sense). No matter what strategy player A chooses, if player B employs the mixed strategy

that is composed with ( 93 x, 69 y), it is guaranteed for him to obtain 183.33. Therefore,

The optimal strategy for player A ( 59 1st , 49 3rd )

The optimal strategy for player B ( 39 x, 69 y)

160

- arfimaHochgeladen vonرضا قاجه
- Insurance Risk and RuinHochgeladen vonIlassa Savadogo
- Probability ExercisesHochgeladen vonAnu Uy
- R Hussain_1aHochgeladen vonawais262
- Unsupervised Learning of Object Deformation ModelsHochgeladen vonFred Funk
- 1-D Spectral Contour Coding For Content-Based Medical Image RetrievalHochgeladen vonInternational Organization of Scientific Research (IOSR)
- Chap 7Hochgeladen vonmeetnaushad
- fifaHochgeladen vonPunit Sarda
- Review of Use of Nonlocal Spectral – Spatial Structured Sparse Representation for Hyper spectral Imagery RestorationHochgeladen vonAnonymous 7VPPkWS8O
- Data MiningHochgeladen vonRobert Doan
- Power-Law Process (PLP) (Gau Yan Xie)Hochgeladen vonSalsacero
- Test of Goodness of FitHochgeladen vonMadanish Kanna
- 2008 Straub ADK Seismic FragilityHochgeladen vonwacotc2163
- Chapter 5Hochgeladen vonAhmad Iskandar Kamarulzaman
- DownloadHochgeladen vonAlexandre Drhdrhrdh
- BLIND SPEECH SEPARATION OF MOVING SPEAKERS IN REAL REVERBERANT ENVIRONMENTSHochgeladen vonda_eusunt
- lecture5 (1)Hochgeladen vonNidhi Ahluwalia
- Chap12Hochgeladen vonSanjayPatel
- Logit to Probit to LPM ExampleHochgeladen vonluke
- CDA OddsRatios 20160121Hochgeladen vonSaleema Karim
- 4. IJAERD - Design and Evaluation of Quick Switching System With Single SamplingHochgeladen vonTJPRC Publications
- Geor IntroHochgeladen vonSílvio De Castro Silveira
- R Tutorial 22 Bayesian Poisson InferenceHochgeladen vonFilone
- Anatomy Pareto Chart 2005Hochgeladen vonsarav.karthik
- 1-s2.0-S1110256X15000747-mainHochgeladen vonDaay Vlix
- Efieciencia parametrosHochgeladen vonJuan123xyz
- Nwobi2014Hochgeladen vonNeng Iyani Maulida
- 26_FRM1_SANDEE_final_report_dilhani.docxHochgeladen vonWirdaAndani
- output-logistic-regression-1.pdfHochgeladen vonValdi Dwirama
- Maximum Likelihood Estimation in the Weibull Distribution Based on Complete and onCensored SamplesHochgeladen vonRaul Diaz

- Week2-Review1Hochgeladen vonkenny013
- from youtube - fast car tabsHochgeladen vonkenny013
- Uchiage Hanabi DAOKO.pdfHochgeladen vonkenny013
- Surface TensionHochgeladen vonkenny013
- Prime Factor Tree Worksheet 50 WITHOUT CIRLCESHochgeladen vonkenny013
- Harmer 2009 Review is Pages About the Pension SystemHochgeladen vonkenny013
- Module 4 Annotated 3141Hochgeladen vonkenny013
- 34235yHochgeladen vonkenny013
- Week 1 and 2 Summary ReadingsHochgeladen vonkenny013
- 15_GIT_1Hochgeladen vonkenny013
- Problem Set 3 (Duration II) With AnswersHochgeladen vonkenny013
- Lecture 5 Foreign Exchange RiskHochgeladen vonkenny013
- Tute SolutionsHochgeladen vonkenny013
- Ch20-3635Hochgeladen vonkenny013
- Bill RatesHochgeladen vonkenny013
- Fins3635 Formulas 2017Hochgeladen vonkenny013
- 16 ACTL2131 ExercisesHochgeladen vonkenny013
- Unit01_FinancialPlanningAndTheProfession2643Hochgeladen vonkenny013
- Assessment RubricHochgeladen vonkenny013
- Guest Lecture Actuarial April 2017 (Rebecca Hartcourt)Hochgeladen vonkenny013
- Kyle Landry M and S Theme.pdfHochgeladen vonkenny013
- Multi-state Actuarial Models of Functional Disability - Adam ShaoHochgeladen vonkenny013
- Module 4 SlidesHochgeladen vonkenny013
- Module 6 ExercisesHochgeladen vonkenny013
- Kyle Landry M and S ThemeHochgeladen vonkenny013
- NQF Resource 02 Guide to ECS Law RegsHochgeladen vonkenny013
- Accounting Framework; AmendmentHochgeladen vonkenny013
- TestHochgeladen vonKevin Jacobson

- 05Continuous Univariate Distributions, Vol. 1.pdfHochgeladen vonUsherAlexander
- Comparison of ruin probability approximations in case of real dataHochgeladen vonnlrlibrary
- HW4Hochgeladen vonJesse Cortez
- Bouchaud-Theory of Financial RiskHochgeladen vonRobin Elocin
- 96510302 vaznoHochgeladen vonEmir Vejselovic
- Modeling Claim Sizes in Personal LineHochgeladen vonTanadkit Munkonghat
- 16 ACTL2131 ExercisesHochgeladen vonkenny013
- ECONOMETRICS ACT601Hochgeladen vonZaheer Ahmad
- Lecture Notes in Artificial IntelligenceHochgeladen vonRobert Peterson
- UECM3463May16-C01(2ph)Hochgeladen vonLim Yen Wen
- QCI VersionHochgeladen vonozy05
- 1465651129 TQM - 601 Module 10- Quality Tools - Pareto ChartHochgeladen von512781
- Lecture-4Hochgeladen vonJoe Ogle
- NoisyCache_ITC25Hochgeladen vonImam Adhita Virya
- fExtremesHochgeladen vonCarlos Eduardo Matos
- Practical Bayesian Model Evaluation Using Leave-One-out Cross-Validation and WAIC - Vehtari Gelman Gabry 2016Hochgeladen vonjohseb71
- Ch 2Hochgeladen vonCsb Finance
- l MomentsHochgeladen vonJavier Senent Aparicio
- Loss Aggregate Part3Hochgeladen vonBlimundo
- exercise-estimation-2017-10-09-13-01-03.pdfHochgeladen vonEric
- Edu 2009 Fall Exam c TableHochgeladen vonTrever Grah
- Sip TesterHochgeladen vonGuru Balaguru
- lossdistHochgeladen vonbudi626
- 1205.1002.pdfHochgeladen vonfredsv
- B.Y.K.(Sinnar) College of Commerce NashikHochgeladen vonsharmasanjiv
- fExtremes.pdfHochgeladen vonthaasha
- Probabilistic Modeling of Extreme Events Using Asymptotic AnalysisHochgeladen vonMitchell Falcon
- Ahp Matrix CsiHochgeladen vonMuhammad Rifqi
- K-th Upper Record Values from Power Lomax DistributionHochgeladen voninventionjournals
- ECQ manual.pdfHochgeladen vonLeonardo Andres Ortega Ochoa

## Viel mehr als nur Dokumente.

Entdecken, was Scribd alles zu bieten hat, inklusive Bücher und Hörbücher von großen Verlagen.

Jederzeit kündbar.