Sie sind auf Seite 1von 411

ii£T¥~1f WWW.GFEDU.

NET

_t ~

=rt CFA,

JiX.JL I¥J -=t¥- ~J} ~tJltiJ,

lfi:J~ ~ffi:&ii~:tif-1J11 ~ ~

I~\

-=t¥- ~

FRM., RFP., CFRM., SIr.,

t§=*31LMflfiJili}jU~,

31LM!I'X~ (1iJFJE1:A~~iJ\) 1tfnf~YU)xxJM%~m, !20TI!lJiJT7F:

T tal%# ~l~ ~ 1f) fik ~ ~ Jl)C¥D ± ~;tJl;tJ;] ttt It ~1f JA ~ :&ii:k tiflJ II, 31L ii~i&£mtlJ ~ ii:ktgf 110~ ~ ~ ±~ 'E1
( ~
Q

tD1i::~pgiJll~

~~~,U~~~*~-~~~~~~o~ffl~*~~~~~,~~~*.~
~ ii~ ~mfo 5ltr,Ii:k~-~AJ ~ oX i~1pJBt@till, fit fI'J ~t t1-'f1~ ftk:i ~fR 8~~ ~ -ifilUj ~ ~ B~PX;)0JJT-=Pi; CFA ~llL iITEr« FRM ~9.L -=.-=P$:g5ltr,li:l!lfJt~~_I%B"Jf1)jt9Z_ i& JlTl f3Ur~~ , ~~f~kE)Jx; J)] 7J )~:& ilE 3'f ' ~ 111~iT l~,T, -*~ ~$ t i A~ ~y:1n\<fH~0
Q

flt11'J ~)('J CF A

to FRM 1* *Er-H*~U fffiM 7'1 £:UW3t ~ !S--1-ti $:>1<-, /f> ItJT±fu ll tJf E

~,A~~~~W0~,~~~~~~m, ~*~.fi~~0~,mfimm*fi ~;tJLitJitt1;j:t*JUt*iJlli5UJL ii£l'E~ In:A$*§A~fIHj\ -liE3t5t~JiJT, iXE3t0~, ¥~0~,~~0~,~~~W0~,~ffi~~0~,:A~~~~~~~~~~ 1¥J~~~~Aln,~liM~~~*,~OO~~M~m~Q~ln~~~~m,~~


±-tbJR, tJT]OlBC, 1tOO, ~OOtD~OO~~~C
± JC!.i'l
0

~%j B~PX;A31L;:ktifiJll¥~~fJ~8~t5.F

iitf~fr f& ~I~ jJ\ ,,~. ~71z § a -5-J a Er0 ~A" B~flit, ~ t~~t tl'fr ~ ~ )] fO 1iJFE /f> J itt~ fit 1I'J § ~ fO~:z; 1I']:g: P M-=f1i£ ~ I¥J :£lffl~, ~ ~1l*f Er>Jt.mflT to {£ fI\J !~'m, ~, itt 1ft ./~--1fd!B~~ Wl~ jHR, 31L iJII, 1i£ Wl~ -i1rJ~O~ ii~~ &t; l!£ ff l;J:g: F 1fr ~ ii~tit f§ 1iJb ~ rPJ B~-17-~Jblilimt'f, ~fZ*~Y;fIlj§.fD iii ~A~~ruj§., MA 1'ij:--·-1fl.:g: F pJ ~ 1~nji:ff!B~ m;J~~
J~\

~**

M-17-%o

~f1!1~1f m. GFEDU.

NET

jg~1fJl¥J~~1JtliH!:JTB<J~~~~iR" ~~~8"J1fr: F ~Jgi)II"~7.k f8"J~~~t;§if3J5fIJ$ ~ e"J~iM!F=b, $_ttft1fJe"J~~jd)()\., J ilft1i'J B<J~ iJH1IiaFfm~1fI~F ~ ~~IJ
~~¥tE~~:it l;J~~7J!=f ILl,

'*

~~~ffl~~~,ffl~~~~~*.ili~~~~~M~~B<J~~~~,~~ ~~~ftm~~~~~;ili+*~~~~~~m~M~~B<J~*,~~~M1fr:~ ~~~~~~~~M~B<J~~~ffl.o~~~~~M~~B<Jm~~1fr:~~~M ~"~~~~"AA~~~B<J~~~*o

www.gfedu.net www.gfedu.org www.frmchina.com www.51jrlk.com www.widepro.org www.51rfp.com www.siichina.com

Book.net

11

~~¥~1f

\'IVIVI.GFEDU. NET

CONTENTS

STUDY MODULE I, QUANTITATIVE ANALYSI8


PART

1
1
J

I PROBABILITY

The Probability of Variables Joint Distribution Expected Value Variance Covariance and Correlation Chebyshev's Inequality Uniform Distribution Binomial Distribution Normal and Lognormal Distribution Poisson Distribution

4 6 6 8 15 15 17 18 25

PART

II STATISTIC

THEORY

27
27 29 29 31 , 32

Skewness and Kurtosis Sample Variance Standard Error of the Sample Mean Confidence lnterval Type I and Type Il

Regression Coefficient of Determination The Test for the Coefficient

40

45
49

PART
VaR

III

QUANTIFYING

VOLATILITY IN VAR MODEL

51
5] 51

Fat Tail EWMA GARCH Implied Volatility Time Conversion Mean Reversion Nonsynchronous Data ;

52 53 55 55 57 58

3;:f¥~W

WlVW. GFEDU. NET

PART IV

Me

SIMULATION AND EVT

59

STUDY MODULE II MARKET RISK MANAGEMENT


PART I FIXED INCOME SECURITIES
Bond Accrued Coupon Strips Yield Curve Curve Fitting
DVOI

65
65
65 71 72 72 74 75 78 90 93 97 99 101

Duration Modified Duration Convexity Callable & Putable Bond Convertible Bond MBS

PART II FORWARD, FUTURE,FRAAND


Forwards

SWAPS

107
107 111 113 , 121 122

Backwardation and Contango...................................................................................................... Futures Margins Forward Rates

PART III OPTIONS


Options Cap and F1oor Put Call Parity The Greek Letters Trading Strategies Swaption Exotic Option ,

142
142 157 158 160 168 171 175

PART IV HEDGING STRATEGIES


Hedging Strategies for Fixed Income Securities Hedging strategies Using Future, Forward and Swap Contracts Hedging Strategies Using Option Contract
II

181
181 183 194

~f¥~1f

WWW.GFEDU.NET

PARTVVAR
VaRProperties VaR Calculation VaR Approaches Simulation Approaches CFaR Backtesting Scenario Analysis Stress Testing Extreme Value Theory (EVT)

198
198 211 227 239 240 ,' 241 242 243 248

PART

VI RISK

BUDGET

250
250 252 255

Risk Budgeting for Pension Funds and Investment Managers Using VaR ISDA Liquidity risk

STUDY MODULE III INVESTMENT AND RISK MANAGEMENT


PART PART

261
261 268
268 275 275 276

I PORTFOLIO MANAGEMENT

IIHEDGE

FUND MANAGEMENT

Individual Hedge Fund Strategies Funds of Hedge Funds Benchmarking Hedge Fund Performance Style Drifts: Monitoring, Detection and Control

STUDY MODULE IV CREDIT RISK MANAGEMENT


PART


279
279
279 281 289 293 298

I DEFAULT RISK

Credit Rating Loan Recovery Rates and Recovery Functions Credit Portfolio Model Credit at Risk and Others

PART

II COUNTERPARTY

RISK

301
III

~l1I1~1f
PART

WWW.GFEDU.NET

III

TRADITION

CREDIT RISK MANAGEMENT

ApPROACHES

....

312
312 313

Netting Securitization

PART

IV CREDIT DERIVATIVES
:

317
317 324 326 327 333

Credit Default Swaps Toatal Return Swap Credit Spread Synthetic Products Subprime Mortgage Credit

STUDY MODULE V OPERATIONAL RISK MANAGEMENT


PART

335
335
335 347 355

I DEFINITION AND MEASURING OPERATIONALRISK

Operational Risk Identification Operational Risk Measurement Operational Risk Management.

PART PART

II

CASE STUDIES FmMWIDE RISK MANAGEMENT

364 369
369 374

III

Firmwide Risk Management Report of the CRMPG II

STUDY MODULE VI

CAPITAL ALLOCATION

375

PART PART

II REGULATORY III

CAPITAL

381

FINANCIAL CONGLOMERATE

403

IV

~f~~1f

WWW.GFEDU.NET

STUDY MODULE I, QUANTITATIVE ANALYSIS

Part I Probability

The Probability of Variables 1. A German Bank lends 100 million DEM to a Russian Bank for one year and receives 120 million DEM worth of Russian government securities as collateral. Assuming that the I-year on the Russian government securities is 20 million DEM and the Russian bank's I-year probability of default is 5 percent, what is the German bank's probability of losing money on this trade over the next year? A. Less than 0.05%. B. Approximately 0.05%. c. Between 0.05% - 5%. D. Greater than 5%.

Answer: C Assuming the VaR is computed at the 99 percent level, there is a 1 percent chance that the collateral value will fall below the value of the loan over the next year. If the probability of default is independent of the decline in the value of the collateral securities, there is a 0.05 percent (= 0.01 x 0.05) chance that the bank will default and the securities will fall below 100 million DEM. On the other hand, if the securities' value falls when the Russian bank defaults, the probability of the German bank losing money is simply the 5 percent default probability. 0n a multiple choice exam with four choices for each of six questions, what is the probability that a student gets less t an two questions correct simply by guessing? A. 0.46% B. 23.73% C. 35.60% D. 53.39% Answer: D The number of questions correct follows a binomial distribution where the probability of success is (1/4) and the number oftrials is 6. Therefore, the probability of getting zero correct, p(O),and the probability of getting one correct, p(1), are: p (0) = (3/4)6= 0.17798 p(l)= 6x(1/4)x(3/4)5 =0.3596 2.

And so, the probability of getting less than two questions correct is p(O) + p(l) = ~ ll::,1( FlI=i )J ('or 01 f!;!:A

~fj1~1f
53.39%.

WWW.GFEDU.NET

Assume you have a portfolio of 10 obligors that are not correlated. Given that each obligor's probability of default is 5 percent, what is the probability that no defaults will occur over the next year? A. 5.0%. B. 50.0%. C. 60;0%. D. 95.0%. Answer: C The probability that each of the 10 obligors will survive without defaulting is (P1 )(P2)' .. (PIO) , where 'p' is the probability of not defaulting

3.

= 1 - default

probability. So the probability that all will survive = (1 - 0.05)1°=60%. 4. Given the below data for the US dollar and Canadian dollar exchange rates, which of the following statements is true? E[CADIUSD Rate] and E[USDICAD Rate] denote the mean of the CADIUSD Rate and USDICAD Rate, respectively. Level USDICAD CADIUSD 1.0000000 1.0 Current 1.1 0.9090909 Up 1.1111111 Down 0.9 Mean 1.0 1.0101010 A. E[CADIUSD Rate] = l/E[USDICAD Rate] B. E[USDICAD Rate] >= lIE[CADIUSD Rate] C. E[USDICAD Rate] <= l/E[CADIUSD Rate] D. E[CADIUSD Rate] = E[USDICAD Rate]

Answer: B As 1.0101010 is greater than or equal to 111.'Ilhis is due to Jensen's Inequality, which says that for any convex function the expectation of [(x) is greater than or equal to the function evaluated at the expectation of x. 5. On any given day, there is a 40% chance that the stock of Company XYZ will go up, a 40% chance that it will go down, and a 20% chance that it will stay the same. A trader buys the stock on the open and sells it on the close on 3 consecutive trading days. What is the probability that he will come out ahead on this game (end up with more money that he started out with)? Ignore interest and transactions costs. A. 0.0604. B. 0.5. C. 0.2080. D. Cannot be determined from the information provided.

:>i£l~$J~ WWW.GFEDU.NET

Answer: D It is impossible to determine an expected return without knowing the magnitude of the up and down movements. Obviously, if the size of up and down moves were the same, the expected return would be zero. 6. A company has a constant 30% per year probability of default. What is the probability the company will be in default after three years? A. 34% B. 44% C. 68% D. 90%

Answer: C 1-(1-0.3)3 7.

= 66%

The annual incremental probability of default of a bond is 15% in year 1 and 20% in year 2. What is the probability of the bond surviving (i.e. no default) to the end of two years? A. 68% B. 65% C. 80% D. 85%

Answer: A Probability (no default) = 8.

(1- o.is)« (1- 0.2)=

0.68 = 68%

A portfolio of bonds consists of five bonds whose default correlation is zero. The one-year probabilities of default of the bonds are: 1%, 2%, 5%, 10% and 15%. What is the one-year probabilify of no default within the portfolio? A. 71% B. 67% C. 85% D. 99%

Answer: A Probability (no default) = (1- 0.01)x

(1-

0.02)x (1- O.OS)x (1-

o.ro). (1-

0.1S)= 0.71 = 71%

9.

The characteristic function of the product of independent random variables is equal to the: A. square root of the product of the individual characteristic functions. B. product of the individual characteristic functions.
C. exponential root of the product of the individual characteristic functions.
3

~l3[~1f WWW.GFEDU.NET

D. sum of the individual characteristic functions. Answer: B Product of the individual characteristic functions The characteristic function of the sum of independent random variables is equal to the product of the individual characteristic functions. E (XY) = E(X) x E(Y). 10. An investor is choosing one of twenty securities. Ten of the securities are stocks and ten are bonds. Four of the ten stocks were issued by utilities; the other six were issued by industrial firms. Two of the ten bonds were issued by utilities; the other eight were issued by industrial firms. If the investor chooses a security at random, the probability that it is a bond or a security issued by an industrial firm is: A. 0.80. B. 0.70. C. 0.60. D. 0.50. Answer: A Let B represent the set of bonds and I the set of industrial firms. The desired probability is the probability of the union of sets B and I, P (B I). According to the theorems of probability, PCB I)=P(B)+P(I)-p(BnI), where PCB) is the probability that a security is a bond=P(B)=l 0/20, P(I) is the probability that a security was issued by an industrial firm=P(I)=14/20, and p(BnI) is the probability that a security is both a bond and issued by an industrial firm=P(BnI)=8/20. 10/20+14/20-8/20=16/20=0.80.

Joint Distribution 11. Calculate the probability of a subsidiary and parent company both defaulting over the next year. Assume that the subsidiary will default if the parent defaults, but the parent will not necessarily default if the suosidiary defaults. Also assume that the parent has an I-year probability of default of 0.50 percent and the subsidiary has a J-year probability of default of 0.90 percent. A. 0.450%, B. 0.500%. C. 0.545%. D. 0.550%. Answer: B Since the subsidiary will default if the parent defaults, t he joint default probability is the 0.5 percent probability that the parent will default.

3I£f~#&:1f WWW.GFEDU.NET 12. A portfolio consists of two (long) assets £100 million each. The probability of default over the next year is 10% for the first asset, 20% for the second asset, and the joint probability of default is 3%. Estimate the expected loss on this portfolio due to credit defaults over the next year assuming zero recovery rate. A. £30 million B. £32 million C. £36 million D. £66 million Answer: A The expected loss for the portfolio is the sum of the estimated loss for each investment (the loss amount multiplied by the probability of the loss).
(0.10)($100 million) +(0.20)($100

million) = $30 million

The joint probability does not affect the expected loss, unless it is in addition to the probability of each event occurring independently. 13. Dependent random variables are defined as variables where their joint probability is: A. equal to zero. B. not equal to the product of their individual probabilities. C. greater than the product of their individual probabilities. D. equal to the product of their individual probabilities. Answer: B Dependence results between random variables when their joint probabilities are not equal products of individual probabilities. If they are equal, then an independent relationship exists. 14. If a random variable X has density f(x) and random variable Y has density g(y), then X and Yare independent of each other if and only if their joint density function h(x,y) satisfies, A. h(x,y)=k f(x)g(y); k;i:l B. h(x,y»f(x)g(y) . C. h(x,y)=f(x)g(y) D. h(x,y)<f(x)g(y) Answer: C 'C' is the definition of independence between two random variables. Two random variables X and Yare independent if and only if their joint density function is the product of the two marginal densities.

~l~~1fWWW.GFEDU.NET

Expected Value 15. If Y = In(X) and Y is normally distributed with zero mean and 2.33 standard deviation. What is the expected value of X? A. 15.10 B. 3.21 C. 227.90

D. 1
Answer: A The expected value can be ca1culated by applying the following equation: 17 I? E(x) = exp(fl+-cr) = exp(O +- 2.33-) = 15.10

Variance 16. Suppose returns are un correlated over time. You are given that the volatility over two days is 1.20%. What is the volatility over 20 days? A. 0.38%. B. 1.20%. C. 3.79%. D. 12.0%. Answer. C The volatility of returns which are uncorrelated from one period to the not can be scaled using the square root of time rule, where the volatility for twenty days is equal to the volatility for two days, times the square root often.

volatility volatility

20

= volatility =1.20%(

(J¥ )

20

J10 )=3.79%

Book.net
{2, 4, 6, 8,

17. What is the standard 10}. A. 2. B. 2.8. C. 3.2. D. 3.6.

deviation of the following set of numbers?

Answer: C The mean of this set is 6. Therefore, the variance is [(6_2)2 + (6_4)2 + (6-6)2 + (6_8)2+ (6-10)2]/4 = 10. Standard deviation = 3.2

3!r:f.'!U~jCj!f WWW.GFEDU.NET

18. Given the following data for a market variable what is the best estimate of its variance? Probability
24% 40% 36%

Value
-12 4

14

A. 6
B. C. D. 10 32 97

Answer: D
First we calculate the Mean = -12 x24% + 4 x 40% + 14 x 36% = 3.76 Then, Variance = 24%(3.76 _12)2+ 40%(3.76 - 4)2 + 36%(3.76 - 14)2= 97.38

19. An analyst is studying a stock that is currently trading at $35. The analyst estimates that there is 33 percent probability that the stock will trade at $50 after one year, a 20 percent probability that the stock will trade at $42, and a 47 percent chance that the stock will trade at $20. What is the volatility of this stock price? A. 13%.
B. C. D. 24%. 31%. 39%.

Answer: D
Step 1. The returns for the three scenarios 20% = (42 35) / 35, and -42.857% Step 2. Calculate expected price = 33% x42.857%+20% x 200/0-fJ-47% x(-42.85'1%)=-2%. Step 3. Calculate Variance = 33% (-2% - 42.857%)2 + 20% (-2% - 20%)2 + 47% (-2% + 42.857%)2 deviation = 39.31 %. given are: 42.857%
=

(50 - 35)/35,

= (20 - 35) / 35

.net

= 0.154538

Step 4. Calculate volatility/standard

20. Use the following joint probability below. Y=1 X=1 0.05 X=2 0.05 X=3 0.15 The expected value ofY is closest to:

distribution Y=2 0.05 0.10 0.15

to answer the questions Y=3 0.10 0.15 0.20

S%:fJE~1f WWW.GFEDU.NET A. 0.2. B. 2.2. C. 1.0. D. 2.3.

If you know that Y is equal to 2, the probability A. 0.05. B. 0.25. C. 0.20. D. 0.17.

that X is equal to 1 is closest to:

The variance of X is closest to: A. 0.54. B. 0.67. C. 0.74. D. 0.61. Answer: B p(Y= 1)=0.05+0.05+0.15=0.25, p(Y=2)=0.05+0.1 0+0.15=0.30, and p(Y=3)=0.1 0+0.15+0.20=0.45, so E(Y)=0.25(1 )+0.30(2)+0.45(3)=2.20. Answer: D
p(X=1IY=2)=0.05/(0.05+0.10+0.15)=0.17.

Answer: D p(X=l )=0.05+0.05+0.1 0=0.20, p(X=2)=0.05+0.1 0+0.15=0.30, and p(X=3)=O.l5+0.15+0.20=.50. Thus, the mean of X is equal to ~=0.20(1)+0.30(2)+0.50(3)=2.3, and the variance is calculated as 0.20(1-2.3)2+0.30(2-2.3 )2+0.50(3-2.3)2=0.61.

Covariance

and Correlation

ook.net

21. Given the following time series of data for x and y, what is the correlation between x and y? x: 10,9,8,7,6,5 y: 5, 6, 7, 8, 9, 10 A. 1. B. -1. C. O. D. 0.5.

~f~~1f WWW.GFEDU.NET

Answer: B There is a one-for-one increase in "y" for every decrease in "x". thus, the correlation is negative one. 22. Given two random variables X and Y, what is the Variance of X given Variance[Y] = 100, Variance [4X - 3Y] = 2,700 and the correlation between X and Vis 0.5? A. 56.3 B. 113.3 C. 159.9 D. 225.0 Answer: D Using the theorems on variance and covariance, Variance [4X-3Y] = 16xVar[X] + 9xVar[Y]

+ 2x4x

(-3) x (Var[X])O.5x (Var[Y]f5xcorrelation[X,Y].

Solve for

Var[X] = 225.0. 23. It has been observed that daily returns on spot positions of the Euro against the US Dollar are highly correlated with returns on spot holdings of the Japanese Yen against the Dollar. This implies that: A. when the Euro strengthens against the dollar, the yen also tends to strengthen against the dollar. The two sets of returns are not necessarily equal. B. the two sets of returns tend to be almost equal. C. the two sets of returns tend to be almost equal in magnitude but opposite in
SIgn.

D. none of the above. Answer: A Correlation describes movement in the same direction but does not imply the same magnitude. 24. If the daily returns of two assets are positively correlated, then: A. the covariance of their daily returns must be positive. B. the covariance of their daily returns must be zero. C. the covariance of their daily returns must be negative. D. nothing can be said about the covariance of their daily returns.

Answer. A If variables are positively correlated, the covariance between the Variables will also be positive. 25. You are given that X and Yare random variables each of which follows a standard normal distribution with Covariance (X, Y) = 0.4. What is the
variance of (5X + 2Y)?
9

:i:~~1f WWW.GFEDU.NET 11.0. B. 29.0. C. 29.4. D. 33.0.


A.

Answer. D Since each variable is standardized,

its variance is one. Therefore, V(5X+2Y) = 25 + 4 + 8 = 37 The closest answer is 33.

25V(X) +4V(Y) +2 x 5x 2x Cov(X,Y)

26. What is the covariance between populations A and B?

A
17 14 12
13

B
22 26 31 29

A. B. C. D.

-6.25. 6.50. -3.61. 3.61.

Answer: A ~ (A - E(A»(B - E(B» The covariance = L.


n=!

Ai

B,

Ai - E(A) +3 ...,...., 0 ~[l-2 -J


I

B/ - E(B) -5 =-1 , +4 C +2

(A-14)(B-27) -15 0 -8 -2 rJ Sum= -25 Avg= -6.25

17 22 14 262 ~ 12 31]_ [29 L 13 Sum=56 Sum=108 Avg=14 Avg=27

et

27. The covariance between variable A and variable B is 5. The correlation between A and B is 0.50. If the variance of A is 12, what is the variance of B? A. 10.00. B. 2.89. C. 8.33. D. 14.40. Answer: C The formula that relates correlations and covariance is:
10

3ltl~~-g

WlVW.GFEDU. NET

Therefore, 0.5 =

.JU*6B

o B = 2.88675 and o ~ = 8.33 28. Which one of the following statements about the correlation coefficient IS FALSE? A. It always ranges from -1 to + 1. B. A correlation coefficient of zero means that two random variables are independent. C. It is a measure oflinear relationship between two random variables. D. It can be calculated by scaling the covariance between two random variables. Answer. B Correlation describes the linear relationship between two variables. While we would expect to find a correlation of zero for independent variables, finding a correlation of zero does not mean that two variables are independent. 29. Given that x and yare random variables, and a, b, c and d are constant, which one of the following definitions is wrong? A.
B.

E(ax + by + c) = aE(x) + bE(y) + c , if x and yare correlated. Var(ax + by + c = Var(ax + by) + c, if x and y are correlated. Cov(ax + by, ex + dy) = acf(ar(x) + bdVar(x) + (ad + be)C'ov(x,y) , if x and y_ are correlated.

C.

D.

Var(x - y)

= Var(x

+ y) = Var(x) + Var(y) , if x and yare uncorrelated.

Answer. B Var(ax + by + c) = Var(ax + by) = a2Var(x) + b2VAR(y) + 2abcov(x, y). 30. The covariance between the return from two securities is 4 and the correlation between them is 0.5. If the variance of the first return is 16, the variance of the second return will be CLOSEST to: A. 0.25.
B. 0.50.
11

~;f¥i¥~1f WWW.GFEDU.NET C. 2.00. D. 4.00. Answer: D Covariance = Correlation x Standard Deviation A x Standard Deviation B. Therefore Standard Deviation ofB = Covariance / Correlation I Standard Deviation A = 4 / 0.5/4 = 2. Therefore Variance of B = 4. 31. A trader in your firm is convinced that the stock index in country X is perfectly negatively correlated to the S&P 500. In order to profit from this analysis, he has taken a long position on index X and shorted S&P 500 futures. Which of the following is TRUE? A. This is a riskless trade because the stocks have negative correlation. B. Apart from the currency risk and cash flow risks on margin calls, this is almost a riskless strategy. C. This trading strategy has the same risk as shorting the S&P 500. This is not a hedged position. D. There will be some small residual risk due to the currency conversion. Otherwise, it is a virtually riskless strategy. Answer: C The long position on negatively correlated index X does not hedge the short S&P 500, but rather the two add to it each other. In order to profit from the negative correlation, one needs to go long on both assets. 32. Asset 1 has correlation of 0.5 with asset 2. A portfolio with equal weights of these two assets has a standard deviation of 13. The standard deviation of asset 2is 19.50. What is the approximate standard deviation of asset I? A. 5. B. 10. C. 20. D. Insufficient Information.

et

Answer: B Let x be the standard deviation of asset 1. Thus 132 = (0.5SDll + (0.5 x 19.5)2 + 2 xO.5 x 0.5 x 0.5 x SD1 x 19.5. Solving this

problem requires trail and error. Always try the middle number because even if it is not the answer, you can judge whether to try a higher number or a lower number. 33. A portfolio contains two perfectly negatively correlated investments with volatilities of 5 percent and 7 percent. The proportion of these two securities that would lead to the lowest risk are:
12

~f!H&:W WWW.GFEDU.NET A. B. C. D. 58% 42% 34% 66% and and and and 42%. 58%. 66%. 34%.

Answer: A Since the investments are perfectly negatively correlated the volatility of the portfolio

= (WI

X Valj

- W2

x var.}. This can be reduced to zero by adjusting the amounts


2=

invested in the inverse of the ratios of the volatilities (i.e. w/w

var./var,

).

34. If Security A and Security B are positively correlated, Security A increases, the price of Security B: A. will increase. B. will decrease. C. is most likely to increase than to decrease. D. may decrease or remain unchanged, but will not increase.

and the price of

Answer: C The positive correlation only means that the prices will tend to move in the same direction, not that they always will. 35. An analyst observes that the closing price of a stock during a week as $33, $43, $45, $48, $46. On the corresponding days the S&P 500 closed at 1150, 1125, 1140, 1160, and 1170. Based on this data the covariance of the stock with the market is CLOSEST to: A. 15.8 B. 18.0 C. 37.4 D. 54A Answer: B Step l. Calculate the mean stock price = (33 + 43 + 45 + 48 + 46) 15 = 43. Step 2. Calculate the mean S&P level = (1150 + 1125 + 1140 + 1160 + 1170) I 5 = 1149. Step 3. The covariance between the stock and S&P = [(33 - 43)(1150 - 1149) + (43 43)(1125 - 1149) + (45 - 43)(1140 - 1149) + (48 - 43)(1160 - 1149) + (46 - 43)(1170 1149)] 15= 18. 36. The variance of the returns from stock A is 0.018 and that of the market is 0.025. If the covariance between the stock and the index is -0.002, their correlation coefficient is CLOSEST to:
A. -0.23.
13

.EliteBoo .net

1Il:f_!j!_~1f

WWW.GFEDU.NET

B. -0.11. C. -0.09. D. -0.08. Answer: C Correlation coefficient

Covariance I (Standard deviation of stock A x Standard

deviation of market index) = -0.002 I (vlO.018 x vlO:025)

= -0.09.

37. The correlation coefficient for two dependent random variables is equal to: A. the product of the standard deviations for the two random variables divided by the covariance. B. the covariance between the random variables divided by the product of the variances. C. the absolute value of the difference between the means of the two variables divided by the product of the variances. D. the covariance between the random variables divided by the product of the standard deviations. Answer: D the covariance between the random variables divided by the product of the standard deviations. Correlation is equal to covariance divided by the product of the standard deviations. 38. For two (possibly dependent) random variables, X and Y, an upper bound on the covariance of X and Y is: A. a(X)· a(Y).
B. 1.

C. there is no upper bound unless the variables are independent. D. zero. Answer: A Cov(X, Y) ::; a(X) • a(Y).

.net

39. Assume the annual volatility of the market is 20% and a stock's annual volatility is 30%. The B of the stock is 1.2. What are the correlation and covariance, respectively, between the stock and the market? Correlation Covariance A 0.048 0.8 B 0.048 0.8 C 0.8 Cannot be determined with the information given D 0.048 Cannot be determined with the information given
14

iI£~~1f

WWW.GFEDU.NET

row a B. rowb c. rowc D. rowd


A.

Answer: A A is correct.

The calculation is
---j- COV

f3 =

cov(Rs,RM) () var RM

Rs,RM

= f3x

() var RM

cov(R\., RAJ) corr = ---'--"-~a (Rs) x a (RAJ )

f3 = 1.2; a (Rs) = 0.3; and a (RM ) = 0.2


cov( Rs' RM) = 1.2x (0.2)2 = 0.048 0.048 co rr (R~, RM) = . = 0.8 . 0.3xO.2 B is incorrect - the values have been swapped (i.e. are in the wrong places). C is incorrect - there is sufficient information for the covariance to be determined. D is incorrect - the correlation calculation is incorrect and there is sufficient information for the covariance to be determined. Chebyshev's Inequality 40. The probability that an observation lies within 3 standard deviations of the mean for any probability distribution is at least: A. 75%. B. 99%. C. 851%. D. 54%.

ok.net

Answer: C 2 One can use Chebyshev's Inequality to calculate this proportion. 1 - (1/3 ) = 89 percent. Uniform Distribution

41. The random variable X with density function

15

~~t&:1f WWW.GFEDU.NET fora < x <h otherwise b). A. B. C. D. Calculate its mean. (a + b)/2 a-bI 2 a +b I 4 a- bl 4 is said to have a uniform distribution over (a,

Answer: A The mean of a continuous uniform distribution with a lower limit of a and an upper limit ofb is (a + b) I 2. 42. Assume we use a continuous uniform distribution U(O,lO) to generate a series of random numbers. Which of the following statements is correct? The number 5 is likely to be observed much more often than any other number. A. Numbers between 4 and 6 are more likely to occur than numbers between 6 and 10, because the first interval is closer to the center of the distribution. B. Numbers between 1 and 3 are as likely as numbers between 4 arid 6. C. Numbers between 1 and 3 are less likely than numbers between 4 and 6, D. due to the skewness of the distribution. Answer: C The cumulative density function for a continuous uniform distribution U(O,l 0) is: F(x) = 0 for x:::; 0 10-0 10· F(x) = 1 jar x > 10 F(l) = /0 = 0.1; F(x)

=--

x-O

=-=for

0 < x <10

3 F(3) = 1 0= 0.3;

F( 4) = 1~ = 0.4;

F(6) = I~ = 0.6;

F(l 0) =

1~ = 1.0

PC1:::; :::; 3) = F(3) - F(1) = 0.2 x

P( 4:::; x:::; 6) = F(6) - F( 4) = 0.2

P(6:::; x:::; 10) = F(IO) - F(6) = 0.4 The probability of x being equal to any specific value (such as 5) is zero for a continuous distribution. Therefore the correct answer is "c", because the probability of getting a value between I and 3 (0.2) is the same as the probability of getting a number between 4 and 6. 43. The probability of an outcome being between 10 and 30 for a random variable that follows a continuous random distribution over the range of 0 to 40 is closest to: A. 0.2. B. 0.5.
16

~fJI!.~lI
C.

WWW.GFEDU.NET

004. D. 0.8. Answer: B P (10:S x:S 30)

(30-10) I (40-0) = O.S = SO%

Binomial Distribution 44. Given that a stock price rises on two days out of five and falls on three days out of five, what is the probability that it will rise on exactly seven out of the next eight days? A. 0.8%. B. 4.S%. C. 9.0%. D. 12.0%. Answer: A This problem requires the use of binomial probability function,

C;~px(1- p

r-

= C; 0.47 (1- 0.4)1 = 0.8%

45. An analyst has determined that the probability that the S&P 500 index will increase on any given day is 0.60 and the probability that it will decrease is 0040. The expected value and variance of the number of up days in a 5-day period are closest to: A. B. C. D. 3.0 and 1.2. 3.0and1.1. 2.0 and 0.5. 2.0 and 2.1.

Answer: A E(X)=np=S(O .60)=3.0. Var(X)=np( I-p }=5(0.60)(OAO)= 1.2.

.EliteB ok.net

46. If the probability of a manager outperforming the median manager statistic is determined to be 58 percent, the number of quarters over the next three years that the manager is expected to outperform the median manager is closest to: A. S quarters. B. 7 quarters. C. 9 quarters. D. 11 quarters.

17

~llftkW m. GFEDU.

NET

Answer: B If outperforming the median manager is considered a success, and underperforming the median manager is considered a failure, this problem can be solved using binomial probability. The expected value would be the number of periods (12) multiplied by the probability of success (58 percent), which yields approximately (12)(0.58) = 7 quarters. Normal and Lognormal Distribution 47. Which of the following statements are TRUE? I. The sum of two random normal variables is also a random normal variable. II. The product of two random normal variables is also a random normal variable. III. The sum of two random lognormal variables is also a random lognormal variable. IV. The product of two random lognormal variables is also a random lognormal variable. A. I and II only. B. II and III only. e. III and IV only. D. I and IV only. Answer: D Statement (I) is true because if "X" and "Y" are random normal variables, then X + Y = Z is a random normal variable, If "X" and "Y" are lognormally distributed random variables, then In(X) and In(Y) are normally distributed random variables. Given that the sum of two normally distributed random variables is a random normal variable, In(X) + In(Y) must be a random normal variable. It follows then that if In(XY) = In(X) ± In(Y~, the product. XY, must lbea random Icgnormal variable, Thus, (IV) is true. 48. Which of the following statements best characterizes the relationship between the normal and lognormal distributions? A. The lognormal distribution is the logarithm of the normal distribution. B. If the natural log of the random variable X is lognormally distributed, then X is normally distributed. C. If X is lognormally distributed, then the natural log of X is normally distributed. D. The two distributions have nothing to do with one another. Answer: C If a variable is itself lognormally distributed, then the log of that variable is normally distributed.

18

1iif~q&1f 1m.
A. O.

GFEDU. NET

49. What is the kurtosis of a normal distribution? B. Cannot be determined: It depends on the variance of the particular normal distribution considered.

C. 2.
D. 3. Answer: D Anormal distribution has a kurtosis of three. 50. If a distribution with the same variance as a normal distribution has kurtosis greater than 3, which of the following is TRUE? A. It has fatter tails than normal distribution. B. It has thinner tails than normal distribution. C. It has same tail fatness as the normal distribution since variances are the same. D. Cannot be determined.from the information provided. Answer: A A distribution with a kurtosis of greater than three will have fitter tails than a normal distribution. 51. For a standard normal distribution, what is the approximate area under the cumulative distribution function between the values -1 and 1? A. 50%. B. 66%. C. 75%. D. 95%. Answer: B The area under a standard l1QrN1!a] distribution witliin + 1 or - 1 standard deviation is approximately 66 percent of the total. 52. Which of the following exhibit positively skewed distributions? I. Normal Distribution. II. Lognormal Distribution. III. The Returns of Being Short a Put Option. IV. The Returns of Being Long a Call Option. A. II only. B. III only. C. II and IV only. D. I, III, and IV only.

19

~f-~~1fm.

GFEDU. NET

Answer: C A lognormal distribution is positively skewed because it cannot contain negative values. The returns on a long call position cannot be more negative than the premium paid for the option but has unlimited potential positive value, so it will also be positively skewed. 53. For which of the following currencies would it be most appropriate to choose a lognormal interest rate model over a normal model? A. USD. B. IPY C. DEM. D. GBP. Answer: B In 1999 yen interest rates were very low. The lognormal model precludes the possibility of negative interest rates, so it is more appropriate when we're using a model to forecast future interest rates with current rates close to zero. 54. The distribution of one-year returns for a portfolio of securities is normally distributed with an expected value of €45 million, and a standard deviation of €16 million. What is the probability that the value of the portfolio, one year hence, will be between €39 million and €43 million? A. 8.6% B. 9.6% C. 10.6% D. 11.6% Answer: B We can compute how many standard deviations away from the mean each value would be and refer to the cumulative distribution tables to determine what portion of the probability distribution lies under these points. With a standard deviation of ] 6 million, the value of 43 million would be: 43-45 or -0.125 standard deviations from the mean (45). and the value 16 of 39 million would be: 39-450r-0 375 standard deviations from the mean. By referring 16 to the distribution tables, we can ascertain how much of the distribution lies under these points. The area between the mean and 39 is 0.14615; and between the mean and 43 is 0.04975. The difference of 0.096 is the value of the distribution which lies between 39 an'143. 55. If we say that commodity returns follow a lognormal distribution, we mean that over time: A. The natural logarithm of the price is normally distributed. B. The change in the price is normally distributed. C. The change in the natural logarithm of the price is normally distributed over time.
20

~f¥<l&1f

WWW.GFEDU.NET

D. The reciprocal of the price is normally distributed. Answer: C A random variable has a lognormal distribution if its logarithm is itself normally distributed. 56. A stock trading at a price of 90 has a (lognormal) price volatility of 40%. The range of prices covered by a 1 standard deviation move up and a 1 standard deviation move down over one year is about: A. 80 points B. 72 points C. 62 points D. 74 points Answer: B One standard deviation up and one standard deviation down would constitute a range of two standard deviations. If the standard deviation is 40 percent and the mean is 90, the range would be [90-(0.4)(90)] to [90+(0.4)(90)] or 54 to 126, a range of72 points 57. The return on a portfolio is normally distributed with an expected rate of return of 10%, and a standard deviation of 20%. What is the probability that the return will be between 0% and 5%? A. 7% B. 9% C. 11% D. 13% Answer: B We can compute how many standard deviations away from the mean each value would be and refer to the cumulative distribution tables to determine what portion of the probability distribution lies under these points. With a mean 10 and standard deviation of 20, the value of 0 would be (0-10)/20 or -0.5 standard deviation from the mean, and the value of 5 would be (5-10)/20 or -0;25 standard deviations from the mean. By referring to the distribution tables, we can ascertain how much of the distribution lies under these points. The area between the mean and 5 is 0,0987, and 0.1915 between the mean and O. the difference ofO.0928(approximately 9 percent) is the value of the distribution which lies between 0 and 5. 58. For a lognormal variable X, we know that In(X) has a normal distribution with a mean of 0 and a standard deviation of 0.5. What are the expected value and the variance of X? A. 1.025 and 0.187 B. 1.126andO.217
C. 1.133 and 0.365
21

~~~1f

WWW.GFEDU.NET

D.

1.203 and 0.399

Answer: C We can calculate the expected value of x using the following formula: E(X)

=e

,u+.!.a2 2

= LIB
-

And the variance of X can be calculated by: Vex) E(X) = e2,u+2(52 e2J1+o"2 = 0.365

59. The lognormal distribution is A. Positively skewed B. Negatively skewed c. Not skewed, i.e., its skew equals 2 D. Not skewed, i.e., its skew equals 0 Answer: A The lognormal distribution is positively skewed or skewed to the right. 60. Consider a stock with an initial price of $100. Its price one year from now is given by S = 100*exp(r), where the rate of return r is normally distributed with a mean of 0.1 and a standard deviation of 0.2. With 95% confidence, after rounding, S will be between: () A. $67.57 and $147.99 B. $70.80 and $149.20 c. $74.68 and $163.56 D. $102.18 and $119.53 Answer: C Tile 95% confidence interval for r1 is -0.2920 to 0.4920: rl = 0.1- (0.2 x 1.96)

= -9.2920
or

or

rl

= 0.1 + (0.2 x 1.96) = -0.4920


=$163.56

net

The 95% confidence interval for Sl is $74.68 to $163.56:


S,

= $100 x e-O.2920 =$74.68

S1 =$100xe0492o

61. A portfolio has a mean value of $75 million and a daily standard deviation of $4.27 million. Assuming that the portfolio values are normally distributed, the probability of the portfolio value falling below $40 million within the next seven days is CLOSEST to: A. 0.10%. B. 1.00%. C. 5.00%. D. 15.87%.
22

~fjltt-g WWW.GFEDU.NET

Answer: A Given that the daily standard deviation is $4.27 million, the standard deviation over 7 days = $4.27 million

xF7 = $11.29

million. Using this standard deviation, the level

of $40 million is 3.1 standard deviations = (75- 40)/ 11.29 from the mean value. Given that the returns are normally distributed the probability of value falling more than 3.1 standard deviations from the mean value is 0.1% (since 100% of the probability falls between'<t 3.1 standard deviations of the mean). 62. A standard normal distribution has: A. no tails. B. fat tails. c. infinite tails. D. asymmetric tails. Answer: C The tail ends of a normal distribution stretch to infinity, although the area under these tails becomes insignificant past ± 3 standard deviation from mean. 63. Which of the following are characteristics of a normal distribution? I. Skewness equal to zero. II. Mean less than median. TIl. Kurtosis greater than zero. IV. Continuous and unbounded. A. I and IV. B. II and III. C. I, II and III. D. I, III and IV.

Answer: D The characteristics of a normal distribution are: It is a continuous distribution. It is bell shaped. It is symmetrical about the mean. It peaks at the mean expected value. It extends theoretically from negative infinity to positive infinity (the probability asymptotically approaches zero at plus and minus infinity) It has a skewness of zero (i.e. it is symmetric) It has a kurtosis (the level of peakedness) of three. Below three the distribution is platykurtic

00

.net

64. Given two variables X and Y that are lognormally distributed, what is the distribution of X * Y'!

...

23

31ifj~-g m. GFEDU.

NET

A. Normal,
B. Lognormal. C. Exponential D. N one of the above Answer: B If X and Yare lognormally distributed, 10g(X) and 10g(Y) will be normally distributed, which means that 10g(X) + 10g(Y) = 10g(XY) is also normally distributed, implying that XY must be lognormally distributed. 65. Which of the following statements concerning probability distributions is/are CORRECT? I. The mean ofa uniform distribution is (a + b)/2. ll. The variance of a standard normal distribution is 1. ID. Risk analysts would not be able to assume a distribution is normal is kurtosis is equal to 3. IV. For a continuous uniform distribution with upper bound 8 and lower bound 1, the probability of X = 3 is approximately 33 percent. A. I, III, and IV only. B. I, II and III only. c. I and II only. D. II only. Answer: C The mean of a uniform distribution is (a + b)/2. A standard normal distribution has a mean of 0 and a variance of 1. For a continuous uniform distribution, the probability of a variable being equal to a specific value is zero. One of the properties of a standard normal distribution is that kurtosis is equal to 3. 66. Which one of the following statements about the normal distribution is NOT accurate? A. Kurtosis equals 3. B. Skewness equals 1. C. The entire distribution can be characterized by two moments, mean and vanance. D. The normal density function has the following expression: 1 1 7 f(x) = c=; exp[--2 (x- ,u)-]

-v2TC0'2

20'

Answer: B The skewness of the normal distribution is 0, not 1. The kurtosis of the normal distribution is 3, the normal distribution can be completely described by its mean and variance, and the density function of the normal distribution is as shown.
24

~~~~w WWW.GFEDU.NET
67. Which of the following statements is the most accurate about the relationship between a normal distribution and a Student's t-distribution that have the same mean and standard deviation? A. They have the same skewness and the same kurtosis. B. The Student's t-distribution has larger skewness and larger kurtosis. C. The kurtosis of a Student's t -distribution converges to that of the normal distribution as the number of degrees of freedom increases. D. The normal distribution is a good approximation for the Student's t-distribution when the number of degrees of freedom is small. Answer: C The skewness of both distributions is zero and the kurtosis of the Student's t-distribution converges to that of the normal distribution as the number of degrees of freedom increases. Poisson Distribution

68. A call center receives an average of two phone calls per hour. The probability that they will receive 20 calls in an 8-hour day is closest to: A. 5.59%. B. 16.56%. C. 3.66%. D. 6.40%. Answer: A To solve this question, we first need to realize that the expected number of phone calls in an 8-hour flay is A = 2 x 8 = 16. Using the Poisson distribution, we solve for the probability that X will be 20. P(X=x)= -;l,Xe-A

xt

P(X = 20) = 0.0559 = 5.59% 69. If n is very large and p is small, the Poisson distribution approximate the binomial distribution with: A. A=np. B. A=np(l-p). C. A=n/p. D. A=nJP Answer: A
25

may be used to

~f.'i[~1f WWW.GFEDU.NET

'#-illc • f&!Jft • ±~11l:

If n is very large and p is small, the Poisson distribution may be used to approximate the binomial distribution with A=np.

www.E ite o k.net

26

~l¥~W

WlVW.GFEDU. NET

Part II Statistic Theory

Skewness and Kurtosis 70. What can be said about a distribution distribution (given the same variance)? A. It has a positive kurtosis. B. It has a negative kurtosis. c. It has a positive skewness. D. It has a negative skewness. that has fatter tails than a normal

Answer: A A normal distribution has a kurtosis coefficient equal to positive three. Anything greater than three indicates a distribution with fat tails. 71. Which type of distribution produces the lowest probability for a variable to exceed a specified extreme value 'X' which is greater than the mean assuming the distributions all have the same mean and variance? A. A leptokurtic distribution with a kurtosis of 4. B. A leptokurtic distribution with a kurtosis of 8. c. A normal distribution. D. A platykurtic distribution. Answer: D a. Incorrect. A leptokurtic distribution has fatter tails than the normal distribution. The kurtosis indicates the leve of fatness in the tails, the higher the kurtosis, the fatter the tails. Iiherefore, the probabilitji of exceeding a specified extreme value will be higher .b. Incorrect. Since answer fl.. has a lower kurtosis, a distribution witli a kurtosis of 8 will necessarily produce a larger probability in the tails. c. Incorrect. By definition, a normal distribution has thinner tails than a leptokurtic distribution and larger tails than a platykurtic distribution. d. Correct. By definition, a platykurtic distribution has thinner tails than both the normal distribution and any leptokurtic distribution. Therefore, for an extreme value X, the lowest probability of exceeding it will be found in the distribution with the thinner tails. 72. It is often said that distributions of returns from financial instruments are leptokurtotic. For such distributions, which of the following comparisons with a normal distribution of the same mean and variance MUST hold? A. The skew of the leptokurtotic distribution is greater,
B. The kurtosis of the leptokurtotic distribution is greater
27

~J~~1f WWW.GFEDU.NET C. The skew of the leptokurtotic distribution is smaller, D. The kurtosis of the leptokurtotic distribution is smaller Answer: B A leptokurtic distribution is characterized as having far tails. Kurtosis is a measurement of the flatness of a distribution, or how fat its tails are. A distribution with a higher kurtosis has fatter tails. 73. What is kurtosis? What is its role in statistical distributions? A. Kurtosis measures the nature of the spread of the values around the mean. It represents the 4th moment of a distribution. A small kurtosis indicates a sharp peak in the middle of a distribution. A population with high kurtosis is usually called leptokurtic. The kurtosis plays an important role in distinguishing those distributions that place additional probability on larger values. B. Kurtosis represents the 3rd moment of a distribution. A small kurtosis indicates f1atness in the middle of the distribution. A population with low kurtosis is usually called leptokurtic. Skewness (and not kurtosis) plays an important role in distinguishing those distributions that place additional probability on larger values. C. Kurtosis can be verified in the four initial moments of a distribution and measures the mean of a distribution. D. Kurtosis can be seen in the second and fourth moments of a distribution and measures the standard deviation of a distribution. Answer. A Kurtosis measures the nature of the spread of values around the mean (4th moment).The kurtosis measure for a normal distribution is 3. Excess kurtosis IS sometimes used in lieu of the kurtosis itself and is the kurtosis number less 3. 'i74. An analyst collects the data for interest rate expectations. The mean expected rate is 2.5 percent, the lowest expectation is 1 percent and the highest expectation is 5 percent. This distribution is: A. sparse. B. normal. C. skewed. D. abnormal. Answer: C In this distribution, the lowest element is 1.5 percent below the mean whereas the highest element is 2.5 percent above the mean. Thus the distribution is skewed towards the right. 75. Left-skewed distributions exhibit:
28

:>ltf¥~lf WWW.GFEDU.NET A. B. C. D. greater mass to the left of the expected value. greater mass to the right of the expected value. a longer tail to the right of the distribution. greater mass close to the expected value.

Answer: B Greater mass to the right of the expected value Left-skewed distributions are those with a longer tail to the left side of the distribution, and whose mass is to the right of the expected value.

Sample Variance 76. The last three year-end returns for a stock are 5, -2, and 1 percent. Using an arithmetic mean, the sample standard deviation is closest to: A. 3.51%. B. 2.87%. C. 1.33%. D. 3.11%. Answer: A The sample arithmetic mean is: X = (5 - 2 + 1)/3 = 1.33% The sample standard deviation is:
1"1

Z(Xi
S

= \..::i_:..:...l

-xl

n -·1

ok.net
Standard Error of the Sample Mean 77. If the mean PIE of 30 stocks in a certain industrial sector is 18 and the sample standard deviation is 3.5, standard error of the mean is CLOSEST to: A. 0.12. B. 0.34. C. 0.64. D. 1.56. Answer: C
29

~l£~1f WWW.GFEDU.NET Standard error of the mean


=

s / j;;

0.64.

78. If the average return of 60 stocks in a certain industrial sector is 25 percent and the sample standard deviation is 15 percent, standard error of the mean is CLOSEST to: A. 1.76. B. 1.82. C. 1.88. D. 1.94. Answer: D Standard error of the mean = Standard deviation / Fn = 15 /

J60

= 1.94.

79. The mean equity risk premium over a 40-year period is equal to 8.0 percent. The standard deviation of the sample is 12 percent. The standard error of the sample mean is closest to: A. 0.30%. B. 1.90%. C. 1.26%. D. 8.00%. Answer: B Note the size of the sample here is the number of years. ax
=

12/

J40

l.90%

80. A return series with 250 observations has a sample mean of 10 percent and a standard deviation of 15 percent. The standard error of the sample mean is closest to: A. 0.06. B. 15.80. C. 0.95. D. 3.87. Answer: C The standard error of the sample mean is the standard deviation of the sample divided by the square root of the number of observations in the sample. In this case, (15 / ../250)
=

0.95.

81. A sample has the following characteristics

30

31£fj!~1f

WWW.GFEDU.NET

The mean of the sample is 2.5% Standard deviation is 1.5% 400 observations in the sample. Which is the standard error of the mean estimate? A. 0.125% B. 0.088% C. 0.053% D. 0.075% Answer: D Standard Error of mean is defined as (Standard deviation) x (1/ n). Given the data in the question, the standard error of the mean = (0.015)x(l/ 400) = 0.075% Confidence Interval 82. Assume the six-month LIBOR is 6.percent and its annualized volatility is 20 percent. Based on this information, the six-month LIB OR should not exceed which of the following with 95 percent confidence within a year? A. 7.97%. B. 7.60%. C. 7.40%. D. 7.02%. Answer: A There is a 95 percent probability that LIBOR will be less than 6% + (1.65)(0.20)(6%) = 7.98%. Thus, with a 95 percent confidence level, LIBOR should not exceed 7.90 percent. 83. The mean age of the 80 employees in a company is 35 and the standard deviation is 15. Assuming that the ages are normally distributed and using 95 percent confidence, we can say that the employees within: the firm fall between: A. 20.0 and 50.0 years. B. 31.7 and 38.3 years. C. 33.8 and 36.2 years. D. 34.6 find 35.4 years. Answer: B The range of the 95 percent confidence interval

mean ± (1.96 x standard deviation 31.7 and 38.3

/ J;z)

35 ± 1.96 x 15/

J80

= 35 ± 3.3

= between

84. A portfolio has a mean value of $60 million and a daily standard deviation of
~lL*§II=(5}f1D~ffiA
31

1ii:f¥t1I:1f

WlVW. GFEDU. NET

$8 million. Assuming that the portfolio values are normally distributed, the lowest value that the portfolio will fall to over the next five days and within 99% probability is: A. $4.5 million. B. $18.4 million. C. $30.6 million. D. $42.1 million. Answer: B Given that the daily standard deviation is $8 million, the standard deviation over five days = $8 million

xFs

= $17.89 million.

Given that the returns are normally distributed, we know that 99% of the outcomes will be above 2.325 standard deviations below the mean, i.e. above $18.4 million. 85. An analyst collects a sample of 50 PIE ratios of stocks that are representative of the market. The mean PIE of these stocks is 20 and the standard deviation is 8.5. What is the 95 percent confidence interval for the mean PIE of stocks in this market? A. 17.21 to 22.79. B. 17.64 to 22.36. C. 18.02 to 21.98. D. 18.31 to 21.69. Answer: B The 95% confidence x Standard

interval

Mean ± 1.96 x Standard error = 17.64 to 22.36.

Mean ± 1.96

deviation / j;; = 20 ± 1.96 x 8.5 IJ50

Type I and Type II 86. According to the Basel backtesting framework guidelines, penalties start to apply if there are five or more exceptions during the previous year. The Type I error rate of this test is 11 percent. If the true coverage is 97 percent of exceptions instead of the required 99 percent, the power of the test is 87 percent. This implies that there is a (an): A. 89% probability regulators will reject the correct model. B. 11% probability regulators will reject the incorrect model. C. 87% probability regulators will not reject the correct model. D. 13% probability regulators will.not reject the incorrect model. Answer. D The power of the tat refers to the probability of rejecting an incorrect model, which is one minus the probability of not rejecting an incorrect model (a type 2 error). Given
32 ~)Ik* gel"I j[ T'JI'l'J!9:j\ [

3l£~~W

WWW.GFEDU.NET

that the power of the test is 87 percent, the probability of a type 2 error, the probability of not rejecting the incorrect model is 1.0 - 0.87 = 13%. 87. Which of the following statements are NOT true? I. Type I error occurs when the null hypothesis is not rejected when it is actually false. II. Type II error occurs when the null hypothesis is rejected when it is actually true. III. Type I error occurs when the alternate hypothesis is wrongly accepted. IV. Minimizing the probability of Type II error maximizes the power of the test. A. I and II. B. I and III. C. II and IV. D. I, II and IV. Answer: A In hypothesis testing we accept the alternate hypothesis if the null hypothesis has been rejected. Type I error happens if the null hypothesis is rejected when it is actually true. Type II error happens if the null hypothesis is accepted when it is actually false. The power of the test is the probability of correctly rejecting the null hypothesis (when it is false), so minimizing Type II errors would maximize the power of the test. 88. A Type I error occurs A. fails to be rejected B. is rejected when it c. fails to be rejected D. is rejected when it when the null hypothesis: when it is false. is false. when it is true. is true.

Answer: D A Type I error occurs when the null hypothesis is rejected when it is true. A Type 11 error occurs when the null hypothesis fails to be rejected when it is false. Test Statistic 89. The standard deviation of pharmaceutical stocks is same as that of the market. An analyst wants to test whether the mean spending by tourists coming to a holiday resort is equal to or less than $2,000 with a 1 percent level of significance. He finds that the average spending by 16 tourists is $2,200 and the standard deviation of the population is $400. The critical value of the Z statistic for this study is: A. 1.65. B. -1.96 c. 2.33. D. 2.58.
33

1i£*¥~~

WWW.GFEDU.NET

Answer: C Since this is a one-tailed test with a 0.01 significance level the critical Z value is 2.33. 90. At-test is used instead of a z-test when: A. the sample size is small. B. greater accuracy is required. C. the variance of the population is known. D. the standard deviation is larger than the mean Answer: A A t-test is used when the sample size is small and the variance of the population is unknown. 91. An analyst wants to test whether the standard deviation of return from pharmaceutical stocks is lower than 0.2. For this purpose, he obtains the following data from a sample of 30 pharmaceutical stocks. Mean return from pharmaceutical stocks = 8%. Standard deviation of return from pharmaceutical stocks = 12%. Mean return from the market = 12%. Standard deviation of return from the market = 16%. What is the appropriate test statistic for this test? A. t-statistic. B. z-statistic. C. F-statistic. D. Chi-squared statistic. Answer: D Tests of the variance (or standard deviation) of a population require the chi-squared test.

92. Which of the following test statistics is most appropriate for conducting the hypothesis test given below? Ho: variance A = variance B; where the test is based on two random independent samples from two normally distributed populations. A. t-statistic. B. z-statistic. C. F-statistic. D. chi-square. Answer: C The test to compare variances of two normally distributed populations is F-statistic. 93. Which of the following statements is TRUE? A. Statistical significance does not suggest economic significance.
34

1i£f¥~1f WWW.GFEDU.NET B. Economic significance and statistical significance are not related. C. A result that is statistically significant may not be economically significant. D. Economic significance statistical significance refer to the same evidence Answer: C Economic significance and statistical significance are not the same. Economic significance takes into account statistical significance but also a number of other issues (externalities such as transaction costs, regulations etc.) Note: the wording of the choices· is a bit tricky. Choice A is right in some cases, but in many situations statistical significance can suggest economic significance, although it cannot confirm significance. 94. An analyst wants to test whether the mean spending by tourists coming to a holiday resort is greater than $2,000 with a 1 percent level of significance. He finds that the average spending by 16 tourists is $2,200 and the standard deviation of the population is $400. This study is a: A. one-tailed test. B. two-tailed test. C. four-tailed test. D. alternate test. Answer: A The hypothesis test that the analyst will construct for this study is Ho: Spending is equal to or less than $2,000; and Ha = Spending is more than $2,000. Thus we have only one rejection region (if the test statistic is higher than the critical statistic), i.e. this is a one-tailed test. 95. The A. B. C. D. test statistic for an F-test of the equality of two sample variances is the: product of the two sample variances. ratio of the two sam le standard deviations. product ofthe two sample standard deviations. ratio of the two sample variances.

o .net

Answer: D The test statistic for an F-test of the equality of two sample variances is the ratio of the two sample variances.
F52

11"1.

le·2

96. You have collected monthly returns for a mutual fund and want to test the null hypothesis that the standard deviation exceeds the advertised standard deviation of 3.5 percent. The most appropriate test statistic is based on a: A. F-test.
B. t-test.
35

C. z-test. D. chi-square test. Answer: D The chi-square test is used to test hypotheses about a single population variance. 97. If the sample size is greater than 30 and population variance is unknown, the appropriate test for the sample mean is the: A. t-test. B. z-test. C. p-test or F-test. D. t-test or z-test. Answer: D The central limit theorem makes it appropriate to use the z-test with an unknown variance if the sample size is large enough (n ~ 30), regardless of the distribution of the population. Since the t- and the z-distributions converge as sample size increases, either test is appropriate, although thet-test is a more conservative estimate.

Hypothesis Test 98. What is the appropriate alternative hypothesis to test the statistical significance of the intercept term in the folowing regression? y= al + a2 (X) + e A. HA: aj:F 0 B. HA: aj>O C. HA: a2:F0 D. HA: a2>0 Answer: A In this regression, a, is the intercept term. To test the statistical significance means to test the null hypothsis that al is equal to zero :versus the alternative that not qual to zero. 99. An analyst wants to test whether the variance of return from pharmaceutical stocks is different from that of the overall market. For this purpose, he obtains the following data from a sample of 21 pharmaceutical stocks and a sample of 41 stocks that are representative of the market. Mean return from pharmaceutical stocks 8% Standard deviation of return from pharmaceutical stocks = 9.2% Mean return from market stocks = 12% Standard deviation of return from market stocks = 13% Based on this information and a 0.05 significance level: A. there is sufficient evidence for a difference between the vanance of pharmaceutical stocks and the variance of the market stocks. B. there is insufficient evidence for a difference between the vanance of
36

"~~*

fc'1 =J)} R B~:JXi\ T

~f.!E~-g WWW.GFEDU.NET

pharmaceutical stocks and the variance of the market stocks. C. there is sufficient evidence that there is no difference between the variance of pharmaceutical stocks and the variance of the market stocks. D. there is insufficient evidence that there is no difference between the variance of pharmaceutical stocks and the variance of the market stocks. Answer: B This problem requires a two-tailed F-test (with the null hypothesis Ho: variance of pharmaceutical industry returns = variance of the market). From the given data the test statistic = 0.132 /0.0922 = 2. The critical value of the F-statistic (for df (1) =40, df (2) =20 and p=0.025) is 2.29. Since the test statistic is lower than the critical statistic, we cannot reject the null hypothesis. 100.Using a sample size of 61 observations, an analyst determines that the standard deviation of the returns from a stock is 21 percent. Using a 0.05 significance level, the analyst: A. can conclude that the standard deviation of returns is higher than 14%. B. cannot conclude that the standard deviation of returns is higher than 14%. C. can conclude that the standard deviation of returns is not higher than 14%. D. none of the above. Answer: A The required test for testing the variance is the chi -squared test. The test statistic = (n - 1) Sample variance / Hypothesized variance =60 x 0.212 /0.142 = 135. To test whether the standard deviation is higher (Ho: standard deviation is lower than or equal to 14%), the critical value of chi-squared will be 79.08 (using df = 60 and p= 0.05). Since the test statistic is higher than the critical value, the anavst can reject the null hypothesis and conclude mat the standard (leviation ofretums is higher than 14%. 101.An analyst is given the task of determining whether a group of 16 active portfolio managers have achieved a significantly higher performance (using a significance level of 0.05) than the average for all portfolio managers over a certain period. Over the period of the study, the active portfolio managers achieved a mean return of 15 percent. Over the same period the mean return for all portfolio managers was 12 percent and their standard deviation was 8 percent. The correct conclusion from this study is that: A. the performance of active portfolio managers is significantly higher than the average for all portfolio managers. B. the performance of active portfolio managers is rrotsignificantly higher than the average for all portfolio managers. C. the performance of active portfolio managers is significantly lower than the average for all portfolio managers.
37

~~~w WWW.GFEDU.NET
D. the performance of active portfolio managers is the same as the average for all portfolio managers. Answer: B The null hypothesis in this study is that the performance of active portfolio managers <= average for all portfolio managers. The value of test statistic for this, Z = (0.15 - 0.12)/(0.08/

M) = 0.03/

(0.08/4)

= 1.5. Since the analyst is using a

0.05 significance level the critical value of Z is 1.65 (the rejection region is above 1.65). As the test statistic is lower than the critical value the null hypothesis cannot be rejected, i.e. the performance of active portfolio managers is not significantly higher than the average for all portfolio managers. 102.You want to test at the 0.05 level of significance that the mean price of luxury cars is greater than $80,000. A random sample of 50 cars has a mean price of $88,000. The population standard deviation is $15,000. What is the alternative hypothesis? A. The population mean is greater than or equal to $80,000. B. The population mean is less than $80,000. C. The population mean is not equal to $80,000. D. The population mean is greater than is $80,000. Answer: D The alternate hypothesis is the statement which will be accepted if the null hypothesis is proven wrong. Therefore, we make whatever we are trying to test as the alternate hypothesis - in this case that the mean price of luxury cars is greater than $80,000, and the null hypothesis as the opposite (the mean price of luxury cars is less than or equal to $80,000). This problem is a common example of how statisticians establish hypotheses by proving that the opposite (i.e. the null hypothesis) is false.

103.An analyst has constructed the following t-test for a portfolio of financial securities whose returns are normally distributed: Number of securities = 40. Ho: Mean return >= 18 percent. Significance level = 0.1 What is the rejection point for this test? A. 1.304. B. 1.684. C. 2.021. D. 2.023. Answer: A
38

~~)jlf.([W WWW.GFEDU.NET

This is a one-tailed test with 39' degrees of freedom and significance level of 0.1. Looking up the Student's t-distribution fDr df = 39 and p = 0.1, we get the critical value of 1.304 104.An analyst wants to test whether the return from pharmaceutical stocks is less volatile than that of the overall market. For this purpose.-he obtains the following data from a sample of 21 pharmaceutical stocks and a sample of 41 stocks that are representative of the market. Mean return from pharmaceutical stocks = 8% Standard deviation of return from pharmaceutical stocks = 10% Mean return from market stocks = 12% Standard deviation of return from market stocks = 13% What is the value of the test statistic for this test? A. 1.30. B. 1.50. C. 1.69. D. 2.25 Answer: C Tests of differences between variances of two. populations require the F-test. The value of the test F-statistic = Higher variance / Lower variance = 0.132 /0.12 = 1.69. 10S.An analyst wants to test whether the variance of return from telecom stocks is higher than 0.04. For this purpose, he obtains the following data from a sample of 51 telecom stocks. Mean return from telecom stocks = 15% Standard deviation of return from telecom stocks = 24% Mean return from market = 12% Standard deviation of return from market = 13% Based on this information and a 0.05 significance level: A. we can say that the varianc€ of telecom firms is lower than 0.04. B. we can.say that the variance of telecom firms is higher than 0.04. C. we cannot say that the variance of telecorn firms is lower than 0.04. D. none of the above. Answer: B Tests of the vanance of a population require the chi-squared test. For this data, chi-squared = (n - 1) Sample variance / Hypothesized variance = 50 x 0.242 /0.04 = 72. Since the

t n

analyst wants to. show that the variance is more than 0.04, this will be chosen as the alternative hypothesis and the null hypothesis will be that the variance is lower than or equal to 0.04. The critical value of the chi-squared statistic (for df=50 and p=0.05) is 67.505. Since the test statistic is higher than the critical statistic, we can reject the null hypothesis (variance <= 0.04), and accept the alternative hypothesis (variance> 0.04).
39

~f~$&1f WWW.GFEDU.NET 106.An analyst wants to test whether the standard deviation of return from pharmaceutical stocks is lower than 0.2. For this purpose, he obtains the following data from a sample of30 pharmaceutical stocks. Mean return from pharmaceutical stocks = 8%. Standard deviation of return from pharmaceutical stocks = 12%. Mean return from the. market = 12%. Standard deviation of return from the market = 16%. What is the critical • value of the statistic for this test at a 0.05 level of significance? A. 17.71. B. 28.81. C. 42.56. D. 45.72. Answer: A Tests of the variance of a population require the chi-squared test. Since the analyst wants to show that the standard deviation is less than 20%, this will be chosen as the alternative hypothesis and the null hypothesis will be that the standard deviation is higher than or equal to 20%. Thus using a probability in the right tail of 95% and degrees of freedom of29, the critical value from the chi-squared table is 17.71. 107.An analyst believes that hedge funds have significantly (using a significance level of 0.05) outperformed the S&P 500 over the past five years. So she picks a random group of 15 hedge funds and finds that their mean return over this period is 85 percent and their standard deviation is 45 percent. During the same period the S&P 500 has risen by 75 percent. The critical value of the t-statistic for this study is: A. 1.21. B. 1.65. C. 1.76. D. 1.96. Answer: C Since this is a one-tailed test with a 0.05 significance level and 14 degrees offreedom, the critical value from the Hable is l.76.

k.n t

Regression

108.Under what circumstances could the explanatory power of regression analysis be overstated? A. The explanatory variables are not correlated with one another. B. The variance of the error term decreases as the value of the dependent variable increases.
40

~*-~~wWWW.GFEDU.NET
c. The error term is normally distributed. D. An important explanatory variable is omitted that influences the explanatory variables included, and the dependent variable.
Answer: D The regression could show a high correlation between two variables, positing that a high degree of the movement of one variable can be explained by movement of the other variable, when movements in both are explained by their relationship to a variable that has been omitted. For example, the values -of variables A and B can both be correlated with the value of C. If C is omitted from the analysis, it will appear that movements in A can be explained by B (or vice versa). 109.Many statistical problems arise when estimating relationships using regression analysis. Some of these problems are due to the assumptions behind the regression model. Which one of the following is NOT one of these problems? A. Stratification B. Multicollinearity C. Heteroscedasticity D. Autocorrelation Answer: A Stratification is not related to regression analysis. Choices B, C, and D describe situations that can produce inaccurate descriptions of the relationship between the independent and dependent variables. Multicollinearity occurs when the independent variables are themselves correlated, Heteroscedasticity occurs when the variances are different across observations, and autocorrelation occurs when successive observations are influenced by the proceeding observations. UO.An analyst is seeking to generate a simple linear regression equation that will allow him to estimate lie returns of a beverage company's stock against the returns of the S&P. 500. The analyst has compiled the following information:
CTS&p 500

= 16.4 percent
company

abeverage

= 22.0

percent

p = 0.82 mean annual return S&P 500 = 8.2 percent risk-free interest rate = 4.0 percent What is the slope coefficient of the regression equation? A. 1.85. B. 1.10. C. 2.34. D. 0.80.

41

~fj~W

WWW.GFEDU.NET

Answer: B This question is tricky because it gives a lot of extra information. The slope of the regression equation is beta. Beta = covariance (x,y)/variance (x). Since we are given the correlation coefficient, we can calculate the covariance as 0.82 x 0.164 x 0.22 = 0.02959. The independent variable here is the S&P 500. The variance of the independent variable is the square of the standard deviation = (0.164)2 = 0.0269. The slope coefficient of the regression equation is (0.02959/0.0269) = 1.10. 111.A factor analysis of the dividend-adjusted returns of ABC Ltd.'s stock price was undertaken to determine which economic factors contributed to its performance. The regression was performed on 460 observations. The results are as follows: Table 1: Predictor Coefficient Standard Error of Coefficient Intercept -.0243 0.005772 .0256 0.017655 All - share - index Industrial index .0469 0.006398 .0012 0.001412 Financial index Table 2: 12,466.47 Sum of Squared Regression (SSR) 1,013.22 Sum of Squared Errors (SSE) Sum of Squared Total (SST) 13,479.69 Which one of the following options correctly describes which variables are significant at the 5% level, and the R2 statistic, respectively? Significant Variables at 5% level R2 statistic A. Intercept; Industrial_index 0.924834 B. Intercept; Industria _index 0.075166 0.924834 C. All_share _ index \ Industrial_Index D. All_share_index; Industrial_Index 0.075166 Answer: A The following table shows the test statistics for eachof the four variables, calculated by dividing the variable coefficient by the standard error. The variable is significant if the absolute value of the t-stat is greater than the critical value from the student's t-distribution for 456 degrees offreedom (which is very close to the z-statistic since the number of observations is so high), i.e. 1.96. Predictor Intercept All - share - index. Industrial index Financial index.
42

k. net

T-stat -.0243/0.005772 = -4.21 025610.017655 = 1.45 .0469/0.006398 = 7.33 001210.001412 = 0.85 ~ il:~( §1"15} 1"101t5i: A.

Significant Yes No Yes No

~f*~1f

WWW.GFEDU.NET

The R2 statistic 0.924834.

IS

defined

as the ratio

of SSR/SST

12,466.47113,479.69

112.A risk analyst performs a simple linear regression on return data comprising three variables evolving in time and obtains, amongst others, the foHowings are the statistics:
Coefficients Intercept 49.94 -38.79 -431.75 -70.40 Standard Error 2.85 138.93 170.50 121.06 t -statistic 17.53 -0.28 -2.53 -0.58

X Variable 1 X Variable 2 X Variable 3

Based on these data at a 95% confidence level, the analyst should conclude that: A. The intercept and IIX Variable 211 are statistically significant B. IIX Variable I" and "X Variable 3" are statistically significant C. "X Variable 111, "X Variable 2" and "X Variable 3" are all statistically not
significant D. More information meaningful is required, such as the corresponding may be made. p-values, before any deductions

Answer: A
A is correct. B is incorrect. (Relatively) - (Relatively) small standard errors and high t-stats are one indication large standard errors and low t-stats are one indication of statistical insignificance. That is, of of indicate statistical indicate statistical D is incorrect. significance. significance. t-stats are not an indication are redundant information if the t-stat is provided.

C is incorrect. Negative

The p-values

the p-values tell one nothing more than the t-stats do.

Using the following ANONA table for Two Questions Below: Sum of Squares (SS)
Regression error 1025 925

DF
5 25

113. The number of sample observations in the regression estimation and total sum of squares (TSS), respectively, are closest to:
Observations A. C. 30 31 TSS 100 1950 1950
43

B. 30

~fj!~1f

WWW.GFEDU.NET

D. 31 Answer: C TSS df =25+5=30. Since TSS df =n-l, SST=1025+925=1950.

100 the number of observations


IS

31.

U4. The R2 and the F-statictic, respectively, are closest to: R2 F-statictic A. 53% 1.1 B. 47% 1.1 c. 53% 5.5 D. 47% 5.5 Answer: C R2=ESS/RSS=1025/1950=53% F= ESSI df RSSI df 102515 =205/37=5.5 925/25

Using the following information for Two Questions Below: Mutilple Regression was used to explain stock returns usmg the following variables: Dependent variable: RET= annual stock returns (%) Independent variables: MKT= Market capitalization/$1.0 million IND= Industry quartile ranking (IND=4 is the highest ranking) FORT=Foliune 500 firm, where (FORT=1 if the stock is that of a Fortune 500 firm, FORT=O if not a Fortune 500 Stock) The regression results are presented in the tables below: Coettciem
Intercept rl)'1arketCapltafizaton lndustry Ranking
Fortune 500

Standani Error t~stalistlc


1. 2100

0.5220
0. GNGe

0.4300 3.0900 2.6100 1.7000 F 12.1000

0.6810 0.0210 O.
(41)1)

0.0150
f).

O. 71l!l2 0.9000 df 3
6 9

2725

O. 5281

0.1390 Significance F 0.0060

Al\!O\lA Regression

SS 20.5969 3. 4031 24.0000

lIAS
6.8656 0.5672

Error Total

US. Based on the result in the table, which of the following most accurately represents the regression equation? A. 0.43+3.09(MKT) + 2.61(IND) + 1.7O(FORT) B. 0.681+0.021(MKT) + 0.04(IND) +0. 139(FORT) C. 0.522+0.046(MKT) + 0.7102(IND) + 0.9(FORT) D. 1.21+0.015(MKT) + 0.2725(lND) +0.5281(FORT) Answer: C
44

:3i£l!E!J&1f

WWW.GFEDU.NET

The coefficient column contains The regression parameters. 116.The expected amount of the stock return attributable to it being a Fortune 500 stock is closest to: A. 0.522 B. 0.046 C. 0.710 D. 0.900 Answer: D The regression equation is 0.522+0.046(MKT) + 0.7102(IND) + 0.9(FORT). The coefficient on FORT is the amount of attributalbe to the stock of a Fortune 500 finn. 117.Which of the following situations is not posiblefrom the result of a mutiple regression analysis with more than 50 observations? R2 Adjusted R2 A. 71% 69% B. 83% 86% C. 54% 12% D. 10% -2% Answer: B Adjusted R2 must be less than or equal to R2. However, ifR2 is low enough and the number of independent variables is large, adjusted R2 may be negtive.

Coefficient of Determination

118.Consider the following estimated linear regression model: Y = 0.08 - O.SX + e

k.net

You also know that the standard deviation of the independent variable is 0.4, and the variance of the dependent variable is 0.09. What is cov(X,Y)? A. 0.01152 B. 0.0288 C. 0.0768 D. 0.096 Answer: D The Cov(X, Y) can be ascertained with the given information by the following: R2 also equals p2, so the correlation coefficient Using this value we can solve for the covariance:
45

p =0.8

:sltfj[Wtlf m. GFEDU.

NET

p = cov(x,y)
(J' x(J'

or 0.8 = COv(x,y) 0.4 x 0.3

or cov(X,Y)=0.096

119.The result of the linear regression is: Y = 0.10 - 0.50 X with a correlation coefficient R = (-0.90). The fraction of the variance of Y attributable to X is equal to: quantities A. (-0.90) B. (+0.90) C. (+0.81) D. (-0.50) Answer.C R-squared is the square of the correlation coefficient and measures the fraction of the variance of Y that is attributable to X. R2 = (-0.90)2 = 0.81 120.1f the correlation coefficient of a linear regression is 0.6, the percentage of variation of the dependent variable that is not explained by the independent variable is CLOSEST to: A. 36%. B. 40%. C. 60%. D. 64%. Answer: D The percentage of explained variation = Correlation coefficient
2

= 0.6

= 0.36 or 36%.

Therefore, the unexplained variation is 64%.

ook.net

121.An analyst is using a statistical package to perform a linear regression between the risk and return from securities in an emerging market country. The original data and intermediate statistics are shown on the right. The value of coefficient of determination for this regression is CLOSEST to: A. 0.043. B. 0.084. C. 0.916. D. 0.957. Risk % (X;) 1.1 1.5
46

retum% (~) 3.2 3.5

~l¥W(~ 2.2 3.6 4.3 5.1 6.7

WWW.GFEDU.NET

4.1 4.5 4.8 5.1 5.2

X = 3.5,Y = 4.343
n
11

L(X;_X)2

=24.9,L(X;-X)~=9.08
1=1

L(~_Y)2

=SST=3.617

L(P'
;=1

11

_y)2

= SSR = 3.312

Answer: C Coefficient of determination = SS RJ SST = 3.312/3.617

= 0.916.

122.Paul Graham, FRM® is analyzing the sales growth of a baby product launched three years ago by a regional company. He assesses that three factors contribute heavily towards the growth and comes up with the following results:
y= b + 1.5 Xl + 1.2X2 + 3X3

Sum of Squared Regression [SSR] = 869.76 Sum of Squared Errors [SEE] = 22.12 Determine what proportion of sales growth is explained by the regression results. A. 0.36 B. 0.98 C. 0.64 D. 0.55 Answer: C Coefficient of Determination i.e. R2 explains pf(J)portionof variation explained By the regression. R2 = SSR/SST, SEE = (SSE/(n-2) )112 , SST = SSR + SSE. Therefore, SSE = 489.29, SST =1359.05, R2 = 0.64 123.A simple linear regression of a stock's returns on an industry index provides the following results: Standard Error Coefficient 2.66 Intercept 4.2 0.65 Industry Index 2.4 Sum of Squares Regression 998.56 Error 180.37 Total 1,178.93 Assume that the sample has 8 years of quarterly observations. Which of the following statements regarding the interpretation of the regression
47

ii£W~1J WWW.GFEDU.NET

is (are) CORRECT? I. The coefficient of determination is 84.7 percent. II. The industry index coefficient is significant at the level. III. The correlation coefficient between the stock's returns industry index is 0.42. IV. The correlation coefficient between the stock's returns industry index is statistically different from zero at the level. A. II and IV only. B. I and II only. C. I, II, and IV only. D. I only.

99 percent

confidence

and the return on the and the return on the 99 percent confidence

Answer: C The coefficient of determination, or R2 = (SSRISST) = (998.56/ 1178.93) = 84.70%. For simple linear regression, we can find the correlation coefficient, r, by taking the square root of R2 = ~0.847 = 0.92. We can test the significance of the correlation coefficient by finding the t-statistic = [0.92~(32-2)] / ~(l-0.847) = 12.88. In general, a t-statistic larger than 3 will be significant at the 99 percent confidence interval for a reasonable number of observations (more than 12). Here we have 32 observations so our degrees of freedom is 32 -1 - 1 = 30. We can also test the significance of the industry index coefficient by taking (2.4/0.65) = 3.69 and see that this is above 3, which is significant at the 99 percent confidence level. 124.An analyst is seeing to predict the returns on the stock of Hirauye Inc, a Japanese conglomerate using the MSCI EAFE index. The analyst has compiled the following information: R(Hirauye) = 5.6 + 1.8X R2= 0.64
(]'2

(MSCI EAFE) == 5.9'6 percent

Book.net

125.Given this information, which of the following statements is/are CORRECT? I. The standard deviation of returns for Hirauye Inc. stock is 54 percent. II. The covariance of returns between Hirauye Inc. stock and the MSCI EAFE index is 0.8. III. The variability of Hirauye Inc. stock explains 36 percent of the variability in the MSCI EAFE index. A. I, II, and III only. B. II and III only. C. I only. D. I and III only. Answer: C
48

~;f_U~1f

WWW.GFEDU.NET

~~

• g!j£z •

:tim

An R-squared of 0.64 would be correctly interpreted that the variability of the MSCI EAFE index explains 64 percent of the variability ofHirauye Inc. stock. Because this is a simple linear regression, the correlation of returns is -/ R2 = r = 0.8. To solve for the standard deviation of returns for Hirauye stock, we can use the formula r = covariance/foic-). We know that the variance of returns of the MSCI EAFE index is 5.76%, which means the standard deviation = O"j = -/0.0576 = 24%. To compute the covariance, we have the beta in the regression equation and the variance of the EAFE index. Since Beta = covariance (x,y)lvariance (x), we can solve for the covariance as 1.8 x 0.0576 = 0.10368. We can then solve for the standard deviation of Hirauye Inc. stock as 0.8 = 0.10368 I (0.240"2)= 0.1920"2= 0.10368; 0"2 0.54. =

The Test for the Coefficient

126.If the F-test shows that the set of X variables explain a significant amount of variation in the Y variable, then: A. Another linear regression model should be tried. B. A t-test should be used to test which of the individual X variables, if any, should be discarded C. A transformation of the Yvariable should be made. D. Another test could be done using an indicator variable to test the significance level of the model. Answer: B Because we are dealing with a set of x variables (a multivariate regression analysis), an f statistic would describe the relationship between the set and the dependent variable. without identifying specific variables in the set. The t-test can be used to determine the relationship of each vaniable in the set to the dependent variable. I2l.An analyst regresses the returns of 60 stocks in a stock market and finds that the best fitting line is: Return = 8% + 9% * Beta If the standard error of the estimate is 6% and the standard error of the coefficient of Beta is 4%, the test statistic for the coefficient is CLOSEST to: A. 1.33. B. 1.43. C. 1.50. D. 2.25. Answer: D The null hypothesis for this test is that the coefficient is equal to zero. Therefore the test statistic = (Observed value - 0) I Standard error = (9% - 0) / 4% = 2.25.
49

128.In an ordinary least squares regression, t-tests are used to determine the statistical significance of: A. the individual parameter estimates. B. the regression. c. a set of parameters. D. the error term. Answer: A T-tests are used to determine the statistical significance of individual parameter estimates in an ordinary least squares regression.

www.Elite

ook.n . t

50

~*j~W WWW.GFEDU.NET Part III Quantifying Volatility in VaR Model

VaR 129.Bank Omega's foreign currency trading desk is composed of 2 dealers; dealer A, who holds a long position of 10 million CHF against the USD, and dealer B, who holds a long position of 10 million SGD against the USD. The current spot rates for USD/CHF and USD/SGD are 1.2350 and 1.5905 respectively. Using the variance/covariance approach, you worked out the 1 day, 95% VAR of dealer A to be USD77,632 and that of dealer B to be USD27,911. If the correlation coefficient between the SGD and CHF is +0.602 and assuming that these are the only trading exposures for dealer A and dealer B, what would you report as the 1 day, 95% VAR of Bank Omega's foreign currency trading desk using the variance/covariance approach? A. B. C. D. USD 97,027 USD 105,543 USD 113,932 Cannot be determined due to insufficient data

Answer: A Note that the question asks for the VAR number to be expressed in USD. Therefore, the first step is to convert the foreign currency positions in terms of USD.Dealer A's position in USD: 10,000,00011.2350 = USD8,097,166 Dealer B's position in USD: 10,000,00011.5905 = USD6,287,331 Given that the VAR of dealer A is USD77,632, we first work the daily volatility for the USD/CHF,denoted here by sCHF By definition we get 8,097,166 x 1.645 x sCMF = 77,632 Therefore, sCHF = 77,632/(8,097,166 x 1.645J = 0.005828 or 0.5828% Similarly, the daily volatility for the USD/SGD, denoted here by sSGD is worked out as follows: sSGD = 27,911/(6,287,331 x 1.645) = 0.002699 or 0.2699%. By definition, the standard deviation of the change in the portfolio which comprises of both the currency pairs over a l-day period is given by:
[(0.005828x8,097,166)2 + (0.002699x6,287,331)2 + 2 x 0.602 x (0.005828 x 8,097,166) x (0.002699 x 6,287,331)]°5= 58,983.

Therefore, The l-day, 95% VAR is 1.645 x 58,983 = USD97,027

Fat Tail

130.In the presence of fat tails" in the distribution of returns, VaR based on the
51

:3Itl~~W

WWW.GFEDU.NET

delta-normal method would (for a linear portfolio): A. underestimate the true VaR. B. be the same as the true VaR. C. overestimate the true VaR. D. cannot be determined from the information provided. Answer: A The VaR would be underestimated because of the greater frequency of losses in the tails of the distribution.

EWMA 131.You need to update a daily volatility forecast using the RiskMetricsTM exponential method with a decay factor of 0.97. Yesterday's forecast of standard deviation was 1 percent. Given that you just observed a return of 2 percent, what will be the new forecast of standard deviation? A. 1.030%. B. 1.044%. C. 1.970%. D. 1.977%. Answer: B The formula for the weighted average forecast for the time "t" variance using the RiskMetrics™ approach is: hi = Ahr_1 + (1 A
-e-

Kr

H)

2,

where "r" is the return for time

t - 1, h.; and hi are forecast variances at time t - 1 and "t", respectively, and 2 is an exponential deny factor. Using the information provided, the updated forecast for standard deviation is:

(hl)T

= [(0.97XO.OIJ2+(Q -0.97)(0.02)2]~

= 1.044%.

132.Until January, 1999 the historical volatility for the Brazilian Real versus the US dollar had been very small for several years. On January 13th, Brazil abandoned the defense of the currency peg. Using the data from the close of business on January 13th, which of the following methods for calculating volatility would have shown the greatest jump in measured historical volatility? A. 250 day equal weight. B. Exponentially weighted with a daily decay factor of 0.94. C. 60 day equal weight. D. All of the above. Answer: B
52

~~~~w WWW.GFEDU.NET
The exponentially weighted moving average (EWMA) puts greater weight on the most recent data. (This is the method used by Risklvletrics, also with a decay factor of 0.94.) 133.The current estimate of daily volatility is 1.5 percent. The closing price of an asset yesterday was $30.00. The closing price of the. asset today is $30.50. Using the EWMA (Exponentially Weighted Moving Average) model (with lambda = 0.94), the updated estimate of volatility is: A. 1.5096. B. 1.5085. C. 1.5092. D. l.5083. Answer: A h, ~ (0.94) (0.015)
2

+ (1- 0.94)

[1n(~~:~)

r~

0.015096 ~ 1.5096%

GARCH

134.Which of the following functional forms corresponds to the GARCH (1, 1) model? Assume that uo, aI, ~l are always positive.

C. D.

)
l-

= ao + at£l-l

al = ao + f3t al_1

I
£ 2 1-1

Book net
. an d a 2 1-1 are t h e previous d ay s re t urns
I

Answer: A The conditional variance at time "t" is expressed by the GARCH model as:

at 2_ a 0 +aj -

£ 2H

h + f3 1a 2 t-l, were

and conditional variance, respectively.

135. The GARCH model is useful for simulating asset returns. Which of the following statements about this model is FALSE? A. The Exponentially Weighted Moving Average (EWMA) approach of

RiskMetrics is a particular case of a GARCH process.


~i[t*g:187tTLTDq19:A 53

:&~tp!(~

WWW.GFEDU.NET

B. The GARCH allows for time-varying volatility. C. The GARCH can produce fat tails in the return distribution. D. The GARCH imposes a positive conditional mean return. Answer: D The GARCH model allows for time-varying volatility by describing the conditional variance as a function of the previous period's volatility and the most recent variance estimate:

Where:
w= rVL
V
=
L

l-a-jJ'

a + jJ + r

=1

a + jJ < 1

It is useful in simulating leptokurtic return distributions with fat tails. The EWMA is a special case of the GARCH model with y = 0, a = 1 ,- y, and P = y. The model does NOT impose the requirement of a positive conditional mean return. 136.Which of the following GARCH models will take the shortest time to revert to its mean? A. ht = 0.05 + 0.03 r 2 I-I +0.96 B. ht = 0.03 + 0.02 r-'_1
?

hI-J

+ 0.95

i:

C. ht = 0.02 + 0.01 r-'_1


? ? D. ht = 0.01 + 0.01 r-I_1

+ 0.97 h'_l + 0.98

i.:

Answer: B A Incorrect. The model that will take the shortest time to revert to its mean is the model with the lowest persistence defined by at
d-

p. In

this case the pe~'sistence

factor is the second largest: a1 + p = 0.03 + 0.96 = 0.99. B. Correct. The model that will take the shortest time to revert to its mean is the model with the lowest persistence defined by a, + p. In this case the persistence factor is the second lowest: aJ + p = 0.02 + 0.95
=

0.97.

C. Incorrect. The model that will take the shortest time to revert to its mean is the model with the lowest persistence defined by a] + factor is the largest: al + P = 0.01 + 0.97 = 0.98.,

p. In this

case the persistence

:1itf¥~1f WWW.GFEDU.NET D. Incorrect. The model that will take the shortest time to revert to its mean is the model with the lowest persistence defined by factor is the lowest:
(Xl (Xl

+ p. In this case the persistence

+ P = 0.01 + 0.98 = 0.99.

Implied Volatility 137.With all other things being equal, a risk monitoring system that assumes constant volatility for equity returns will understate the implied volatility for which of the foUowing positions by the largest amount: A. Short position in an at-the-money call B. Long position in an at-the-money call C. Short position in a deep in-the-money call D. Long position in a deep in-the-money call Answer: D A plot of the implied volatility of an option as a function of its strike price demonstrates a pattern known as the volatility smile or volatility skew. The implied volatility decreases as the strike price increases. Thus, all else equal, a risk monitoring system which assumes constant volatility for equity returns will understate the implied volatility for a long position in a deep-in-the-money call. 138.Equity options exhibit volatility smirk patterns greater for: A. high strike price options. B. in-the-money put and call options. C. low strike price options. B. out-of-the-money put and call options. in which implied volatility is

Answer: C Equity options exhibit the "smirk" pattern, where implied volatility is greater for low strike price options. This would include call options, which are out-of-the-money and put options that are in-the-money. Time Conversion

k.net

139.A small hedge fund is running a portfolio with a 5-(1.,y VaR of $3.1 minion. Assuming normal conditions what is the best estimate for VaR over a 2-day horizon'? A. $1.2 million. B. $2.0 million. C. $2.5 million.
55

~~¥~1f WWW.GFEDU.NET

D. $3.1 million. Answer: B 2-day VaR = 5-day VaRx (2/5)°.5 = $3.1 million x 0.4°·5=$2 million 140.Consider a portfolio with a 1 -day VaR of $1 million. Assume that the market is trending, with an autocorrelation of 0.1. Under this scenario, what would you expect the 2-day VaR to be? A. $2 million. B. $1.414 million. C. $1.483 million. D. $1.449 million. Answer. C The 2-day variance=(a2_day)2 = (a1_day)2(2+2p)li, where p is the autocorrelation. Assuming the same

Similarly, the 2-day VaR is: VAR2_day = VAR1_da/2 + 2p)li.

distribution for the two time horizons, the 2-day VaR for the information provided is: VAR2_day = S1.0M(2.2)12 = 1.483 million. 141.ln comparing the one month ahead volatility forecasts from a standard deviation model with an estimation period of 100 months STD(100)) to that of 50 months (denote STD(50)), you would expect A. STD(50) to be more biased on average than the STD(100) relative volatility B. The standard error of the estimate STD(50) to be higher than the
S'fD(lOO)
)/

simple (denote to true one of

C. Both to be about the same D. Cannot tell

Answer: B A longer estimation period increases the stability of the standard deviation estimate (i.e., reduces the standard enor of the estimate) in a simple moving average model because outlier observations have a lower weight and smaller impact on the estimate. Therefore, the standard error of STD (50) would be higher than that of STD(1 00). 142.To the A. B. C.
56

convert VaR from a one-day holding period to a ten-day holding period VaR number is generally multiplied by 2.33 3.16 7.25

:~~J¥~1fWWW.GFEDU.NET

D.

10.00

Answer: B Square root of lOis 3.16. 143.A commodity-trading firm has an options portfolio with a two-day Valueat-Risk (VaR) of $2.5 million. What would be an appropriate translation of this VaR to a ten-day horizon under normal conditions? A. $3.713 million. B. $4.792 million. C. $5.590 million. D. Cannot be determined Answer: C Using the square root of time formula, the 2-dayVaR could be translated into a l O-day VaR by: $2.5M(lS)

= $5.590million

144.The VaR of a portfolio at a 95% confidence level is 15.2. If the confidence level is raised to 99% (assuming a one-tailed normal distribution) the new value of VaR will be closest to: A. 10.8. B. 5.2. C. 18.1. D. 21.5. Answer: D 95% confidence level requires a volatility multiple (alpha) of 1.65 while 99% confidence level requires a multiple of 2.33. Since YaR is directly proportional to this multiple, the 99% confidence level VaR = 15.2 x 2.J3 ! 1.65 = 21.5

Mean Reversion

145.Unlike stock prices, interest rates appear to be pulled back to some long-run average level. What is the name of this phenomenon? A. Regression B. Mean reversion C. Inversion
57

',." .,~:

ii£f\f~1f WWW.GFEDU.NET

D. Conversion Answer: B Reversion to the mean, or mean revision, describes the tendency of some variables (in this case interest rates) to tend to move toward the mean or long-term average value. 146.Assume we estimate volatility and calculate a one day VaR. Ifvolatility is mean reverting what can we say about the.t day VaR? A. It is less than the B. It is equal to

J1

one day VaR

J1

one day VaR

C. It is greater than the

J1

one day VaR

D. It could be greater or less than the

J1

one day VaR

Answer: D Mean reverting volatility implies that there is an expected change in volatility over the time horizon. A simple mean reversion equation would be: Ds = k(m - s) + e Where k is the speed of mean reversion, s is the volatility, and 11is the mean volatility. The key here is to note that if volatility is greater then the mean then we expect tomorrow's volatility to be less than today's. If however volatility is less then the mean then we expect tomorrow's volatility to be greater than today's. Therefore, 'a', 'b' and 'c' are incorrect answers. Nonsynchronous Data

147.An analyst wants to estimate the correlation between stocks on tlie Frankfurt and Tokyo exchanges. He collects closing prices for select securities on each exchange, but notes that Frankfurt closes after Tokyo. How will this time discrepancy bias the computed volatilities for individual stocks and correlations between any pair of stocks, one from each market? There will be: A. increased volatility with correlation unchanged. B. lower volatility with lower correlation. C. volatility unchanged with lower correlation. D. volatility unchanged with correlation unchanged. Answer: C Volatility will be unchanged as all information is priced into share value on both exchanges, but the correlation will decrease as the closing prices reflect the sum of market information available at different times.
58

~;f!jltQ:-g WWW.GFEDU.NET

Part IV

Me Simulation

and EVT

148.You have implemented the simulation process discussed above using a time interval Lit = 0.001, and you are analyzing the following stock price path generated by your implementation.
t 0

Stock. price at start of period 130:00 530.1 I) )30.26 530.42 330.45 $30.46 nO.67 $30.75 $31.11 nUl

)pmpie value forI' 0.3902 0.6597 0.6539 0.1 065 0.0416


0,26Q]

(h;:mge instod;

d urlll9 period
$(UQ $0.113
I";

S,ockprke at startof period 11 12 B 14 15 16 17 18 19 20

Sample value for" (1,7499 0,1:546 0._9347


03Bl4

Change in stock price during period .$CU.9


S(U)7

c, s

t?1.62

$O.O? $(1.21

4
(1

sO.en so,os
$0,19
$0.\6

7 8 9

10

OJ1 10 0.7786 (1,6466 0)3188 (1,1998

-$31.,86 $31.96 H2,17 $.32J9

$(U4 SO.10

0.8101 0.8518
lU47S

$31.46

$0.21 $(1.05

0.7911 0.9147 05941

$0,07 $0,21 $0,24 SO.16

Given this sample, which of the following simulation steps most likely contains an error? A. Calculation to update the stock price. B. Generation of random sample value for s, C. Calculation of the change in stock price during each period. D. None of the above. Answer: B The validity of the calculation to update the stock price is easily verified, as is the validity of the calculation of the change in the stock price during each period. The fact that after 20 time periods, the sample value of e has never been negative is very suspicious and likely contains an error. 149.Which of the following statements about Extreme Value Theory (EVT) and its application to value at risk are true?
I.

II.

III.

IV.

EVT extends the Central Limit Theorem to the distribution of the tails of independent, identically distributed random variables drawn from an unknown distribution. II. For empirical stock market data, the shape parameter in EVT is negative implying tails that disappear more rapidly than a normal distribution. III. EVT can help avoid a shortcoming of the historical simulation method which may have difficulty calculating VaR reliably due to a lack of data in the tails. IV. For empirical stock market data, standard value at risk estimates at the 95 percent confidence level are exceeded more often than 5 percent of the time and would therefore benefit from the use of extreme value theory. A. I and III
59

~~!£!W:1f m. GFEDU.

NET

B. II and IV C. I, III and IV D. III and IV Answer: A I. Correct. Whereas the Central Limit Theorem concerns the distribution of the average of independent, identically distributed variables drawn from an unknown distribution, EVT deals with the distribution of the tails. II. Incorrect. The shape parameter in EVT for empirical stock market data is typically between 0.2 and 0.4, implying that the tails disappear more slowly than a normal distribution. III. Correct. Due to its reliance on historical data which may lack sufficient tail data (i.e., extreme events), reliably calculating VaR with the historical simulation method can be difficult; EVT can help avoid this shortcoming. v. Incorrect. For empirical stock market data, standard value at risk estimates at the 95 percent confidence level tend to be fairly accurate, and generally only becomes inaccurate at the 99.5 percent confidence level and beyond. I50.The A. B. C. D. generalized extreme value (GEV) distribution is most useful for: VaR calculations Stress testing Determining the actual distribution Identifying structual charges

Answer: B Focusing on the distribution of the maximum makes GEV most appropriate for stress testing 151.EVT seeks to overcome a weakness in VaR caused by the assumption of

A. timited data points B. Fat tails C. Normal distributions D. High loss threshhold

et

Answer: C Because reality does not always follow a normal distribution, a more accurate pictrure of the distribution tail is needed for high cofidence calculations. 152.Which of the following about the most common distribution used for peaks-over-threshhold is FALSE? J) The distribution requires a threshhold, shape, and scaling parameter. II) The distribution of these extreme values follows the GEV distribution III) The distribution produces a curve that dips below the normal distribution prior to the tail and then moves above the normal distribution in a curved
60

~f¥fl:l'i WWW.GFEDU.NET
IV)

A. B. C. D.

shape until it reached the extreme tail The distribuiton provides a more accurate estimate of the event probabilities in the distribution tail, allowing VAR to be computed at high cofidence levels I and II II only III and IV IV only

Answer: B GPD requires a threshhold, shape, and scaling parameter. The distribution of these extreme values follows the GPD. GPD produces a cure the dips below the normal distribuition prio to the tail and then moves the normal distribution in a cured shape until it reaches the extreme tail. GPD provides a more accurate extimate of the event probibilities in the distribution tail, allowing VAR to be computed at high cofidence levels.

153.Which of the following statements is true regarding the bootstrap simulation method used in VaR estimation? l) Bootstraping uses actual market data II) The bootstraping method always uses a time horizon based on the time scale of the historical data III) Bootstraping is based on synthesis of normally distributed random numbers A. I only B. II only C. I and III D. I, II and III Answer: A Bootstraping uses actual market data. Bootstraping can be done with data that uses the same time line as the 0lJ(~ ~)~ interest or shorter term data . .MC is based on a synthesis of normally distributed random numbers. 154.Which of the following statement is FALSE regarding the implemention of Me simulation forVAR? I) MC simulation involves the creation of a distribution of pricing paths created via a random generation procedure II) After samples are made, a pricing model is used to determine financial values III) MC simulation used actual market movments of past price data. A. I only B. II only C. I and III D. I, II and III Answer: C
61

~f¥~1f WWW.GFEDU.NET

Me estimation used actual market movments of past price data.


ISS.Extreme value theory (EVT) can assist with value at risk (VaR) calculations by providing better probability estimates of observing extreme losses than that indicated by a standard normal distribution because empirical distributions exhibit fat tails. If one uses the generalized Pareto distribution (GPD method to generate parameter estimates for the shape parameter, fat tails will indicate a: A. positive parameter estimate and VaR calculations that are too large. B. negative parameter estimate and VaR calculations that are too small. C. positive parameter estimate and VaR calculations that are too small. D. negative parameter estimate and VaR calculations that are too large. Answer: C Fat tails will generate a positive shape parameter, which indicates that VaR estimates are probably too small. IS6.EVT, Extreme Value Theory, helps quantify two key measures of risk A. The magnitude of an 'X' year return in the loss in excess VaR B. The magnitude ofVaR and the level of risk obtained from scenario analysis C. The magnitude of market risk and the magnitude of operational risk D. The magnitude of market risk and the magnitude of credit risk. Answer: A Extreme Value Theory or EVT looks at the value of losses beyond an identified cutoff point. Choice A assumes that the cutoff point is identified using excess VaR. IS7.Many fiancial applications are concerned only with extreme values of returns or exceptional losses for which we use extreme value distributions (EVD). The following is/are example(s) of EVDs: I. II. III. V. Weilbull distribution Frechet distribution Generalized Pareto distribution Student's t distribution A. B. C. D. I and II I, II and III IV only II and III

Answer: B Weilbull and Frechet distributions are examples of Generalized Extreme Value distributions. Both distributions are used to model maximal loss. Generalized Pareto
62

~f¥~1f

WWW.GFEDU.NET

distribution is used to model excess over a threshold. However, Student's t distribution is not an example ofEVD and cannot be used for modeling extreme values.

www.EliteBook.net

63

~~~1fWWW.GFEDU.NET

www.Elit

64

3tf¥~l§r WWW.GFEDU.NET

STUDY MODULE II, MARKET RISK MEASUREMENT AND MANAGEMENT

Part I Fixed Income Securities

Bond

1. In which of the following securities does the coupon income rise when the

interest A. B. C. D.

rates fall? Inverse floater. Coupon floater. Dual-index floater. Deleveraged floater.

Answer: A The coupon of an inverse floater is defined as a constant amount minus a reference rate (k LIBOR). Therefore, this coupon amount rises when the reference interest rates falls. 2. Which of the following statements about standard bonds with no optionality is TRUE? I. Higher coupon implies shorter duration. II. Higher yield implies shorter duration. III. Longer maturity implies larger convexity. A. I and II only. Bo.....III III only. and C. I and III only. D. I, II, and III. fixed rate government

o .net

Answer: D Duration is inversely related to yield and coupon. Convexity is directly related to maturity. 3. In managing a portfolio of domestic corporate bonds, which of the following risks is least important? A. Interest rate risks. B. Concentration risks. C. Spread risks. D. Foreign exchange risks.
65

~f¥~1f WWW.GFEDU.NET

Answer: D Foreign exchange risk is not relevant when managing a portfolio of domestic bonds. 4. The yield of a bond can be defined as: A. another way to represent its trading price. B. the rate ofretum. c. the rate of return only if the horizon is equal to maturity. D. the rate of return only if the horizon is equal to its duration. Answer: A The yield of bond is the discount rate used to establish the price of the bond under the assumption that the bond is held until maturity and coupon payments are reinvested at the current yield Within this context a bond's yield is a common means of representing its trading price. 5. The following Treasury zero rates are exhibited in the marketplace: 6 months = 1.25% 1 year = 2.35% 1.5 years = 2.58% 2 years = 2.95% Assuming continuous compounding, the price of a 2-year Treasury bond that pays a 6 percent semiannual coupon is closest to: A. 105.20. B. 103.42. C. 108.66. D. 105.90. Answer: D
3e(-O.0125 x 0.5)

+ 3e(-O.0235

xl)

+ 3e(-00258

x 1.5)

103e(-00295

x 2)

= 105.90

6. Tile zero coupon bond of an Aerated company maturing III five years is trading at a spread of 1% over the zero-coupon bond of a AAA-rated company maturing at the same time. The spread can be explained by: I. Credit Risk II. Liquidity Risk III. Tax differential A. I only B. I and II only C. I and III only D. I, II, and III Answer: B 'I' is correct. default risk.
66

Part of the spread between A-rated and AAA-rated bonds is related

iitt¥~W

WWlV. GFEDU. NET

'II" is correct. Part of the spread between A-rated and AAA-rated bonds is related to liquidity risk. 'III is incorrect. Zero corporate bonds receive the same tax treatment regardless of the credit rating. 7. A two-year zero-coupon bond issued by corporate XYZ is currently rated A. One year from now XYZ is expected to remain at A with 85% probability, upgraded to AA with 5% probability, and downgraded to BBB with 10% probability. The risk free rate is flat at 4%. The credit spreads are flat at 40, 80, and 150 basis points for AA, A, and BBB rated issuers, respectively. All rates are compounded annually. Estimate the expected value of the zero-coupon bond one year from now (for USD 100 face amount). Fixed Income Securities A. USD 92.59 B. USD 95.33 C. USD 95.37 D. USD 95.42 Answer: C The correct answer is: c. USD 95.37 The expected value of the zero coupon bond one year from now is given by 0.05*100 + 0.85*100 + 0.1*100 =95.37 1+ 0.044 1+ 0.048 1+ 0.055 8. The purchase price of a 3-year 9 percent semi-annual coupon bond that is currently yielding 7 percent will be: A. 105.11. B. 105.25. C. 105.33. D. 105.45. Answer: C Using a bond calculator, N = 6, PMT = 4.5, I%/yr = 3.5, FV = 100. Solving for PV we get 105.33. 9. For purposes of computing the market risk of a U.S. Treasury Bond portfolio, it is easiest to measure: A. yield volatility because yields have positive skewness. B. price volatility because bond prices are positively correlated C. yield volatility for bonds sold at a discount and price volatility for bonds sold at a premium to par. D. yield volatility because it remains more constant over time than pnce volatility, which must approach zero as the bond approaches maturity. Answer: D
67

:31t:;ffi[~1f

m. GFEDU.

NET

Because prices must converge to the bond's maturity value, the volatility of yields provides a more accurate risk assessment. 10. The horizon rate of return from a bond consists of two parts: RR, the reinvestment return and CG, the capital gain (or loss) representing the value of the bond at the horizon. Generally speaking, if you increase the coupon on a bond while holding the yield constant: A. RR increases since cash flow is now larger. B. CGincreases since cash flow is now larger. C. RR decreases since the cash flow is now larger. D. RR and CO are always equaL Answer: A The horizon yield on a bond is a function of capital gains and coupon reinvestment income. Holding all other factors constant, coupon reinvestment income increases as the reinvestment rate increases 11. A fixed rate bond, currently priced at 102.9, has one year remaining to maturity and is paying an 8 percent coupon. Assuming the coupon is paid semiannually, what is the yield of the bond? A. 8%. B. 7%. C. 6%. D. 5%. Answer: D For the l-year bond the yield is: I PV = -102.9; FV = 100; N = 2; PMT = 8; -x 2= 2.4955 x2 = 4.99 percent
y

12. Suppose XYZ Gorp. has two bonds paying semi-annually according to the following table: T-bill rate (bank Remaining Coupon(as 30/360) Price discount) Maturity 6 months 8.0% 99 5.5% 1 year 9.0% 100 6.0% The recovery rate for each in the event of default is 50 percent. For simplicity, assume that each bond will default only at the end of a coupon period The market-implied risk-neutral probability of default for XYZ Corp. is: A. greater in the' first six -month period than the second B. equal for both coupon periods. C. greater in the second six-month period than the first. D. cannot be determined from the information provided
68

:>iif.¥~-g

WWW.GFEDU.NET

Answer: A The first step to answering this question is to calculate the yield on the 6-month bond A future value of $104 in six months, with a present value of $99 results in a yield of 10.1% (annual). The spread over Treasuries is then 4.6 percent. The risk of default in the second six months can be determined by the spread on the I-year bond (3.0%), which is much lower than the spread for the first six months. This is sufficient to answer the question without calculating the actual implied default rates. A larger spread corresponds to a higher implied default probability. 13. A bond portfolio manager invests $20 million in a bond issued at par that matures in 30 years, and which promises to pay an annual interest rate of 9%. The interest is paid once per year, and the payments are reinvested at an annual interest rate of 8%. The first payment is one year from today. What is the annual yield on this investment? A. 8.185% B. 8.285% C. 8.385% D. 8.415% Answer: C The total value of the investment at the end of the 30-year period can be computed and then related to the initial investment of $20 million, solving for the internal rate of retum. Using a financial calculator, the value of the coupon payments in 30 years is: N=30; I1Y=8; PMT=$1,800,000; CPT PV= $203,909,780 The total value of the investment at the end of 30 years, including the par value payment, is $203,909,780+$20,000,000=$223,909,780. The internal rate of return is then: N=30; PV=$20,000,000; FW=$223,9109,980; CPT 1IY=8.385% 14. A firm has just issued $1,000 face value bonds with a coupon rate oRBpercent, paid semi-annually, and a maturity of 15 years. If the issue price for this bond is $785.50, what is the yield-to-maturity, stated annually? A. 9.872 percent B. 10.365 percent C. 10.942 percent D. 11.120 percent Answer: C Relating the cash flows of $40 semiannually for 15 years and $1000 at maturity to the present value of $785.50 results in a yield of 5.471 percent per period Stated annually, this would be 10.942 percent. On the TI BA II Plus financial calculator: N=30; PMT= 80/2=40; FV=I,OOO; PV=-785.50; CPT I/Y=5.471 *2=10.942%.

69

~*j!_~1} WWIV. GFEDU. NET

15. Consider a $1,000 par value bond with a 7 percent annual coupon. The bond pays interest annually. There are 2 years remaining until maturity. What is the current yield on the bond assuming that the required return on the bond is 10 percent? A. 10.00% B. 7.38% C. 5.00% D. 3.52% Answer: B N = 2; FV
CY

1000;

PMT

70;

I1Y

10;

CPT

PV

948.

= $70

$948

0.0738 = 7.38%.

16. Consider a $1,000 par value bond with a 7 percent annual coupon. The bond pays interest annually. There are 2 years remaining until maturity. What is the current yield on the bond assuming that the required return on the bond is 10 percent? A. 10.00% B. 7.38% C. 5.00% D. 3.52% Answer: B N = 2; FV
CY:-:: $70

1000;

PMT

70;

I1Y

10;

CPT

PV

948.

$948

= 0.0738 = 7.38%.

17. In a stable flat yield curve environment, the price of a fixed coupon bond trading at a premium will: A. Fall with time. B. Rise with time. C. Not change with time. D. First rise and then fall with time.

Answer: A In a stable flat yield curve environment, the price of a premium bond will fall, reaching par value at maturity. 18. Assume the prices are for settlement on June 1, 2005, today's date. Assume semiannual coupon payments: Coupon Maturity Price 7.500% 12/1/2005 102-9 12.375% 6/1/2006 107-15
70

~l¥q&1f

WWW.GFEDU.NET

6.750% 12/1/2006 104-15 5.000% 6/1/2007 102-9+ The discount rates associated with the bonds maturing June 2006, are closest to: A. 0.9696/0.9858. B. 0.9858/0.9546. C. 0.9546/0.9696. D. 0.9778/0.9696.

in December 2005 and

Answer: B We must calculate the 6-month discount factor first. This is done by dividing today's price by the final payment's par + coupon:
d1

l +~)
'102 32

= 0.9858

(100 + 7 5)

The 12-month discount factor d2 solves the following equation:

C2.;75
=> d2

xO.9858) + (100 + 12.;75}d2)

107 + ;~

= I] .9546

Accrued Coupon 19. Consider a 7.75 percent semi-annual coupon bond with a par value of $100 and four remaining coupons, which is trading at a yield of 8.375 percent. There are 74 days remaining in the current period that has a total of 182 days. lIhe accrued coupon of this Dond is CU0SEST to: A. 1.59. B. 2.29. C. 3.l8. D. 4.57.

Answer: B Step 1. Calculate w = 74 /182 = 0.41. The accrued coupon = coupon x (1 - w) = 100 x 3.875% x (1 - 0.41) = $2.286

:~J¥t9:1f WWW.GFEDU.NET Strips 20. Which of the following statements with regard to STRIPS are true? I. Interest STRIPS from one bond are fungible with interest STRIPS of other bonds that mature on the same date. II. Principal STRIPS from one bond are fungible with principal STRIPS of other bonds that mature on the same date. III. Interest on STRIPS is reported as income when the security matures or the investor sells it in the market. IV. STRIPS have a lower reinvestment risk than a whole bond A. I and II. B. I and IV. C. II and IV. D. I, II and III. Answer: B Interest STRIPS from one bond are fungible with interest STRIPS of other bonds that mature on the same date, but the principal STRIPS are not. Interest on STRIPS is reported as income as it accrues, irrespective of when the security matures or is sold by the investor. STRIPS are essentially zero coupon bonds, and therefore unlike whole bonds, they have no reinvestment risk, i.e. the risk that the investor may not be able to invest the coupons at a high enough rate until the maturity of the bond

Yield Curve
21. Suppose a risk manager has made the mistake of valuing a zero-coupon bond using a swap (par) rate rather than a zero-coupon rate. Assume the par curve is upward sloping. The risk manager is therefore A. Indifferent to the rate used B. Over-estimating the value of the bond C. Under-estimating the value of the bond D. Lacking sufficient information

Answer: B If the par curve is rising, it must be below the spot curve. As a result, the discounting will use rates that are too low, thereby overestimating the bond value. 22. What is the relationship between yield on the current inflation-proof bond issued by the U.S. Treasury and a standard Treasury bond with similar terms? A. The yields should be about the same. R The yield of the inflation bond should be approximately the yield on the treasury minus the real interest.
72

~lj~1f\~.GFEDU.NET

C. The yield of the inflation bond should be approximately the yield on the treasury plus the real interest. D. None of the above is correct. Answer: D The yield on the inflation-protected expected inflation.

bond is a real yield, or nominal yield minus

23. Which of the following statements about yield curve arbitrage is TRUE? A. No arbitrage conditions require that the zero coupon yield curve is either upward sloping or downward sloping. B. It is a violation of the no-arbitrage condition if the one-year interest rate is 10% or more, higher than the lO-year rate. C. As long as all discount factors are less than one but greater than zero, the curve is arbitrage free. D. The no-arbitrage condition requires all forward rates be non-negative. Answer: D For a no-arbitrage condition to hold, both spot and forward rates must be positive. 24. Suppose that the yield curve is upward sloping. Which of the following statements is TRUE? A. The forward rate yield curve is above the zero-coupon yield curve, which is above the coupon-bearing bond yield curve. B. The forward rate yield curve is above the coupon-bearing bond yield curve, which is above the zero-coupon yield curve. C. The coupon-bearing bond yield curve is above the zero-coupon yield curve, which is above the forward rate yield curve. D. The coupon-bearing bond yield curve is above the forward rate yield curve, which is above the zero-coupon yield curve. Answer: A With an upward sloping curve, the coupon curve is the lowest, the zero-coupon curve is above the coupon curve and the forward curve is above the zero-coupon curve. The order is reversed if the curve is downward sloping. 25. If the yield curve is downward sloping: A. par coupon yields will be higher than zero coupon rates, which will be higher than forward rates. B. zero coupon rates will be higher than par coupon yields, which will be higher than forward rates. C. forward rates will be higher than zero coupon rates, which will be higher than par coupon yields. D. forward rates will be higher than par coupon yields, which will be higher
73

:3Ir:;f¥W:.1f WWW.GFEDU.NET than zero coupon rates. Answer: A Par coupon yields are weighted averages of the zero coupon rates, so in a downward sloping term structure they will be higher than the zero coupon rates for the same term to maturity. The forward rates are equivalent to the zero coupon rates plus the slope of the zero coupon curve, thus they will be lower than the zero coupon rates in an upward sloping term structure. 26. BankAsia uses a simple bootstrapping method to compute the USD zero-coupon yield curve from money market rates, futures prices and swap rates. ABC bank also follows the same method but uses FRA instead of futures prices to build its USD yield curve. If the PVOI of BankAsia's dollar swap book is negative, the mark to market valuation using the yield curve of ABC bank will be: A. marginally higher than using its own yield curve. B. marginally lower than using its own yield curve. C. same as using its own yield curve. D. none of the above. Answer: A The PVOI will be marginally higher, because there is less convexity in FRA prices than there is in futures prices due to the mark-to-market features of futures contracts.

Curve Fitting 27. Given a one-year and a three-year zero coupon bond price of 95.18 and 83.75 respectively, what should be the price of a two year zero coupon bond using linear interpolation on zero rates (semiannual compounding)? A. 95.18. • B. 89.47. C. 89.72. D. 83.75. Answer: C Step one is to determine the semiannual zero rates for the 1- and 3-year bonds. For the I-year bond: PV=-95.18; FV=100; N=2; I -x Zj = CPT I1Y=2.5008; ZI = 5.0% 2 For the 3 l-year bond: PV=-83.75; FV=100; N=6; 1 3Z3 = CPT I/Y=2.997; z. = 6.0% Thus, we know the l-year and 3-year zero rates are 5.0 percent and 6.0 percent,
74

~f_lE~~

WWW.GFEDU.NET

respectively. Next, interpolate using the relationship Plugging in known value: S.O = Z2 12

ZI 11 = Z2 12 = Z3 13

= 6.0/3
(2-1) (3 -1)

So 1ve f,or Z 2 as
6.0-Z2 =112

c.

10

11 (6.0-Z2) ows:

(6-S)

The 2-year bond price is: I/Y=0.S*Z2

= 2.7S; N

= 4; FV = 100; PV

= 89.72.

DV01

28. Roughly, what government bond? A. 70,000,000. B. 700,000. C. 100,000. D. 70,000. Answer: D

is the DV01 of a CAD 100 million lO-year Canadian

The DV01 can be computed as

Dmod

(O.OOOlj x Value. If the CAD bond were a

zero coupon bond, its Macaulay duration would be 10 years. Assuming that the CAD bond is a coupon bond, its duration must De less than 10 years, and its modified duration would be less than 10 years, depending on prevailing market rates. Thus, the dollar duration for the bond in the question must be less than: 10 x (0.0001) x $100,000,000 = $100,000. It is reasonable to assume that the modified duration of a 10-year coupon bond is around seven years, thus $70,000 is a reasonable estimate for the DVOlofthe 10-ycar Canadian government bond 29. Assume that the DVOl of an interest rate swap is proportional to its time to maturity (which at the initiation of the swap is equal to T). Also assume the interest rate curve moves in parallel, is stochastic (with constant volatility), and movements in the rates are normally distributed At what time will the maximum potential exposure be reached? A. T/4.

B. T/3.
75

~W~1f WWW.GFEDU.NET

C. T/2.
D. (3/4)T. Answer: B IfDVOl is proportional to time to maturity, then so is duration (D), so we can write: Dt= k(T-t) where: Dt = duration at time "t" k = constant T= time to maturity Then the swap volatility is: crswap=[k(T t)]cr-vt With a little calculus we can show that the point at which exposure) is maximized, is t = T13 (and, therefore, potential

30. Assuming other things constant, bonds of equal maturity will still have different DV01 per USD 100 Face Value. Their nvnr per USD 100 Face Value will be in the following sequence of Highest Value to Lowest Value: A. Zero Coupon Bonds, Par Bonds, Premium Bonds B. Premium Bonds, Par Bonds, Zero Coupon Bonds C. Premium Bonds, Zero Coupon Bonds, Par Bonds D. Zero Coupon Bonds, Premium Bonds, Par Bonds Answer: B A. Incorrect. Premium Bond will have a higher Base Price and hence higher DVOI than that of Zero Coupon Bond. B. Correct. DVOI is certain multiple of Dirty Price (which includes Coupons) and not Clean Price. Thus, it is proportional to Base Price, which is Dirty Price. Ordinarily, Premium Bond will have the highest (clirty) price followed by Par Bond and with the least price of Zero Coupon Bond. Hence, BVO 1 of Premium Bond is the highest while that of Zero Coupon Bonds is the lowest. C. Incorrect. Base Price of Par Bond is higher than that of Zero Coupon Bond and hence, its DVOI cannot be less than that of Zero Coupon Bond. D. Incorrect. DVOI per USD 100 Face Value is an Absolute Amount ofUSD based on actual Base Price Change. Ordinarily, Base Price of a Zero Coupon Bond will be lower than that of Par & Premium Bond. Hence, DVO1 of Zero Coupon Bond is less than that of Premium Bond of same maturity. 31. Assuming the long-term yield on a perpetual note is 5 percent, compute the dollar value of a 1 bp. change in the yield (DV01 ) for a perpetual note paying a USD 1,000,000 annual coupon. A. -20,000. B. -30,000.
76

jff.1¥t'nr

WWW.GFEDU.NET

C. -40,000. D. -50,000. Answer: C The value of a LISD 1,000,000 consol bond (perpetuity) with a yield equal to 5 percent is 1,000,000 1 0.05 =USD 20,000,000. The duration of a consol bond is: De = 1 + liP, where R is the yield So, De =1 + 1/0.05 =21. DV01 (the dollar value of a . one basi point yie ld ch ange ). IS: DV01 = PxDcxO.OOOl . So, we h ave DVO1 = aSIS . I+R 20,000,000 x 21x 0.0001 . .. = 40,000. Thus, for everyone baSISpoint mcrease (decrease) 1.05 in yields, the value of the consol bond will decrease (increase) by USD 40,000. Implied in this problem is an increase of one basis point. 32. Roughly estimate the DVOI for a 2 x 5 CHF 100 million FRA in which a trader will pay fixed and receive floating rate. A. CHF 1,700. B. CHF (1,700). C. CHF 2,500. D. CHF (2,500). Answer: C The duration of a 2 x 5 FRA is 5 - 2 = 3 months, or 0.25 years. The dollar value of a basis point for a CHF 100 million position is: 0.25 x 100M x 0.0001 = CHF 2,500. 33. Suppose you have a position of $100 million in the instruments below. Each one has a maturity of 10 years. Which instrument is most likely to have a DVOI that exceeds the DVOI of a Treasury strip with Ill-year maturity? A. B. C. Inverse floating rate s@Q;1:lrities. D. Corporate zero coupon notes. Answer: C 34. Given the following portfolio of bonds: Bond Price Paramount held (in USDmillion) A 101,43 3 B 84.89 :;
C

e 00 .net
Modified duration
236 4013

izi.s.

6.21

What is the value of the portfolio's DVOI (Dollar value of 1 basis point)? A.8,019 B.8,294

c. 8,584
77

~l1E~1fWWW.GFEDU.NET

D.8,813 Answer: C The portfolio dollar duration of a basis point CDVO = (portfolio modified duration X 1) market value of portfolio )/10,000 The portfolio modified duration is obtained by taking the weighted average of the modified duration of the bonds in the portfolio. Mathematically, it is as follows: wlDI + w2D2 + W3D3 + ...+ w.D, where
Wi

= market value of bond i/market value of the portfolio

D, = modified duration of bond i K = number of bonds of the portfolio. Based on the above, the market values are as follows: bond A = 101.43 x 3,000,000 100 bond B = 84.89 x 5,000,000 100
=

3042900 "

= 4 244 500 "

bond C = 121.87 x 8,000,000 = 9 749 600 100 " Total market value of the portfolio = 3,042,900 + 4,244,500 + 9,749,600 = 17,037,000 Portfolio modified duration is calculated as follows:
--'--'---x.

3,042,900 17,037,000

236 + 4,244,500 x. 13 + 9,749,600 4 17,037,000 17,037,000

6 ')7 =
.L..

0.1786x2.36 + 0.2491 x4.13 + 0.5723x6.27 = 0.4215 + 1.0289 + 3.5881 = 5.0385 Therefore, the portfolio dollar duration of a basis point CDVO is obtained as follows: 1) (5.0385 xI7,037,000) = 8,584 10,000 .

Duration

35. What is the best estimate of the market value of a portfolio of USD tOO million invested in recently issued 6% to-year bonds and USD 100 million of long to-year zero coupon bond if interest rates decline by 0.50%: A. USD 219 million B. USD 195 rriillion C. USD 209 million D. USD 206 million
78

#'::f.lE~1i

WWW.GFEDU.NET

Answer: C To calculate the best estimate of the market value of the portfolio if interest rates decline 0.5%, one needs to calculate the change in the market value of each bond using duration. The duration of the 10-year zero coupon bond is 10. Thus, the change in value of this bond equals 10 x 0.005 x 100,000,000, which equals 5 million dollars. The duration of the newly issued 6% bond is 7.802 assuming that the price of the bond is par. Given a duration of 7.802, the change in the value of the bond equals 7.802xO.005x 100,000,000 which equals 3.91 million. Thus, the best estimate of the market value of the portfolio if interest rates decline by 0.5% is 200 million + 5 million + 3.91 million which equals 208.91 million. Thus, the correct answer is 'C'. 36. Which of the following is not a property of bond duration? A. For zero-coupon bonds, Macaulay duration of the bond equals its years to maturity. B. Duration is usually inversely related to the coupon of a bond C. Duration is usually higher for higher yields to maturity. D. Duration is higher as the number of years to maturity for a bond selling at par or above increases. Answer: C Duration usually increases as the time to maturity increases (Figure 1-4), so D is correct. Macaulay duration is also equal to maturity for zero-coupon bonds, so A is correct. Figure 1-5 shows that duration decreases with the coupon, so B is correct. As the yield increases, the weight of the payments further into the future decreases, which decreases (not increases) the duration So, C is false. 37. A money markets desk holds a floating-rate note with an eight-year maturity. The interest rate is floating at tliree-tnonth ILIBOR rate, reset quarterly. The next reset is in one week. Wliat is the approximate duration of the floating-rate note? A. 8 years B. 4 years C. 3 months D. 1 week Answer: D Duration is not related to maturity when coupons are not fixed over the life of the investment. We know that at the next reset, the coupon on the FRN will be set at the prevailing rate. Hence, the market value of the note will be equal to par at that time. The duration or price risk is only related to the time to the next reset, which is 1 week here.

79

:G::&~1f WWW.GFEDU.NET 38. Bond A has a price volatility of 21 percent and yield volatility ofH percent and Bond B has a price volatility of 1.4 percent and yield volatility of 12 percent. Given this information, what can one conclude about the durations of these bonds? A. Bond A and Bond B have similar durations. B. Bond A has a longer duration than Bond B C. Bond A has a shorter duration than Bond B D. There is no relationship between their durations. Answer.B Duration is a measure of bond price sensitivity to changes in interest rates. Yield volatility does not playa direct role in the determination of duration. 39. A 10 year zero coupon bond is callable annually at par (its face value) starting at the beginning of year six. Assume a flat yield curve of 10 percent. What is the bond duration? A. . 5 years. B. 7.5 years. C. 10 years. D. Cannot be determined based on the data given. Answer: C The duration of a zero coupon bond is always equal to its maturity, regardless of whether it is callable, If the issuer calls a zero, it will be for the purpose of retiring debt, not refunding at lower rates. 40. A 10 year reverse floater pays a semiannual coupon of 8% minus 6-month LIB OR. Assume the yield curve is 8 percent flat, the current 10-year note has a duration of 7 years and the interest rate on the note was just reset. What is the duration of the note? A. 6 months. B. Shorter than 7 years. C. Longer than 7 years. D. 7 years.

Answer: C If the 6-month LIBOR is 8 percent,the current coupon would be zero and the duration of the reverse floater would equal the duration of the note (i.e., seven years). Assuming a positive coupon, the duration of the reverse floater is greater than the duration of the straight note (i.e., greater than seven years). 41. Which of the following would always shorten a non-callable bond's duration? A. Downgrade in credit rating B. Upgrade in credit rating
80

~f¥~1f

WWW.GFEDU.NET

c. Twist in the yield curve


D. A merger with another firm Answer: A Lower credit rating leads to higher discount rate, decreasing weight at the long end. 42. A portfolio consists of two positions: One position is long $100M of a two year bond priced at 101 with a duration of 1.7; the other position is short $SOM of a five year bond priced at 99 with a duration of 4.1. What is the duration of the portfolio? A. 0.68. B. 0.61. C. -0.68. D. -0.61. Answer: D The modified duration of a portfolio is equal to the market weighted average of the modified durations of the bonds in the portfolio. The portfolio's market value is 101 x$100 million _ 99 x$50 million = 101 - 49.5 = 51.5. Note that the short 100 100 position is negative. The market weighted duration of the long position is:
OJ[ongD]ong

= 101 x1.7= 3.33. Similarly, the market weighted duration of the short 51.5

position is (ushort Dshort =- 49.5 x 4.1 = -3.94. Thus, the duration of the portfolio is 3.33 51.5 - 3.94 =-0,61. The negative duration indicates that the portfolio will lose value when yields decrease. 43. Suppose a 30-year fixeu-rate Hond with a market value of 100 million is split into a floater and an inverse floater with a market value of 80 million and 20 million respectively. Assume tlie floater- has a duration of zero and the 30- year bond has a duration of 8. What is the duration of the inverse floater? A. 8. B. 16. c. 32. D. 40. Answer: D The duration of the bond must be the market weighted duration of the floater and inverse floater components. Since the duration of the bond is given as 8.0, the duration of the components can be expressed as: OJ/loaterD/loCiter

+ OJmverseDinverse

= 8.0,

where the OJ'S and D's indicate weights and duration, respectively. Since the duration of the floater is zero, the duration of the inverse floater component can be calculated
81

~~j~1f

WWW.GFEDU.NET

by solving: 8 = 20 x (Dinverse).Thus, he duration of the inverse floater is 40. t 100 44. Coupon curve duration is a useful method to estimate duration from market prices of an MBS. Assume the coupon curve of prices for Ginnie Maes in June 2001 is as follows: 6 percent at 92, 7 percent at 94, and 8 percent at 96.5. What is the estimated duration of the Ts? A. 2.45. B. 2.4. C. 2.33. D. 2.25. Answer: B Coupon curve duration is a version of effective duration, which is computed as D= (V_- V+) , where

2Vatlr

Va

s the price of the bond for which the duration is being

determined, V_is the value of the bond that has the next higher coupon rate, V + is the value of the bond with the next lower coupon rate, and tl r is the incremental change in coupon rates (usually 0.01). Using the values provided, the duration can be compute d as: D= (96.5 - 92.0) =.2 39 5 6= 2 .4. 2 x94x 0.01 45. A and B are two perpetual bonds, i.e., their maturities are infinite. A has a coupon of 4 percent and B has a coupon of 8 percent. Assuming that both bonds are trading at the same yield, what can be said about the duration of these bonds? A. The duration of A is greater than the duration of B B. The duration of A is less than the duration of B C. A and Booth hav€ tile same duration. D. None of the above. Answer: C The duration of a perpetuity (consol bond) is 1 + 1/R, where 'R" is the yield to maturity. Since both trade at the same yield, they must both have the sine duration. 46. Consider a portfolio consisting of two bonds. One is a 5 NC3 bond with a coupon of 5.63 percent. The other is a bullet bond with 5-year maturity and a coupon of 5.50 percent. Suppose that the volatility of the interest rate increases. If this happens, then: A. the duration of the 5 NC3 will approach the duration of the 5-year bullet bond B. the duration of the 5 NC3 will remains unaffected C. the duration of the 5 NC3 will equal the duration of the 3-year bullet bond
82

ok net

iiifj[;j&1f WWW.GFEDU.NET

D. none of the above. Answer: D If the volatilities of interest rates increase, the duration of the 5 NC3 would decrease but would not equal the duration of the 3-year bullet bond 47. Suppose that the coupon and the modified duration of a 10-year bond priced to par is 6.0 percent and 7.5, respectively. What is the approximate modified duration of a 10-year inverse floater priced to par with a coupon of 18 % -2 x LIBOR (1 month)? A. 7.5. B. 15.0. C. 22.5. D. 0.0. Answer: C Separate the fixed-rate bonds into 2/3 FRN and 1/3 inverse floater. This ensures that the inverse floater payment is related to twice LIB OR. As a result, the duration of the inverse floater must be three times that of the bond, so 3 * 7.5= 22.5. 48. Duration of a fixed rate bond, in the case of flat yield curve, can be interpreted as (where B is the bond price and y is the yield to maturity): A.
---

1 8B

BOy

B.

1 8B B 8y
---

G.

Y 8B Bay

.EliteBook.net
Oy 1

D.

(1+ y) 8B

Answer: A Duration is a measure of interest rare risk, so: 8B 8y =-D*B~D=--

8B 8y

83

~l~~W

WWW.GFEDU.NET

49. Consider the following bonds: Bond Number Maturity (yrs)

Coupon Rate

1
2

10 10 10 10

6% 6%
0%

Frequency 1 2 1 1

Yield (

6% 6% 6%
5%

3
4

6%

5 9 6% 1 How would you rank the bonds from the shortest to longest duration? A. 5-2-1-4-3 B. 1-2-3-4-5 C. 5-4-3-1-2 D. 2-4-5-1-3

6%

Answer: A Duration is the average time to receipt of the cash flows, weighted by the present value of each cash flow in proportion to the total price of the bond Other factors held constant, the higher the yield, the lower PV of a cash flow, therefore, a shorter duration. The greater the frequency of the cash flow, the shorter the duration, because the time to receipt of the cash flows is less. The largest single cash flow is at maturity, and, thus, the maturity has the most effect upon duration. We know bond #3 has a duration of 10 because it's a zero coupon bond In this question, the bond with the shortest duration would be bond #5 because its maturity is shorter than the other 6 percent coupon bonds. The second shortest duration would be bond #2 because its yield is equal to the highest (6 percent), and it has the highest frequency (2) of the 6 percent bonds. Next would be bond #1 because it also has the 6 percent yield but a frequency of 1. Bond #4 would come next, because although it has the same cash flows as bond # 1, these would be discounted at the lower yield of 5 percent. The longest duration would be bond #13, which is a zero-coupon bond 50. Which one of the following long positions is more exposed to an increase in interest rates? A. A treasury bill B. lO-year fixed coupon bond C. 10-year floater D. 10-year reverse floater Answer: D The 10-year reverse floater has the highest effective duration and hence the most exposure to an increase in interest rates. Both the 10-year floater and the T-bill have short durations (bills have maturities of 52-weeks or less, while floaters, and have
84

~~~1f

WWW.GFEDU.NET

average durations equal to the time to reset). The l O-year fixed coupon bond falls in between the bill and the reverse floater

51. What can cause the total duration of a bond index to decrease even as the durations of each subs ector (short, mid, and long) increase? A. There is a coupon payment on the largest bond in the index. B. Some bonds in the index rollover from long to mid, and some from mid to short. C. The whole yield curve shifts down by a substantial amount. D. This cannot happen. Answer: B As a bond rolls from one sub-sector to another, it can increase the duration of each subsector, while lowering the duration for the portfolio as a whole. For example, if the portfolio has a sub-sector of bonds with maturities of 2-10 years, and a bond rolls from die longer maturity sub-sector to the mid-maturity one, it is removing a bond from the longer maturity sub-sector that has z lower duration as compared to others in that sub-sector, and adding it to the mid-maturity sub-sector with a duration value that is greater than the other bonds in the mid-maturity sub-sector. The aging of the bond lowers the duration for the portfolio as a whole. 52. Which of the following fixed-income securities most likely has negative effective duration? A. A range accrual note B. A floating rate note C. An interest-only tranche of a CMO D. A principal-only tranche of a CMO Answer: C The C01'reGt nswer is '(S', an interest-only tranche ofa CMQ transactiefl. a

.net

An 1-0 tranche has negative duration because a decline in interest rates causes the 1-0 price to fall. As rates fall and mortgages begin to prepay, the flows of an 1-0 tranche vanish. (Whenever some of the principal is paid-off there is less available from which to collect interest.) When rates are very high and prepayments are low, the 1-0 is like a security with a fixed set of cash flows.It has greater notional amount based on which interest will be calculated. The price of an interest-only tranche will most likely increase as interest rate increases, leading to a negative effective duration. 53. Calculate the duration of a two-year bond paying a annual coupon of 6% with yield to maturity of 8%. Assume par value of the bond to be $1,000: A. 2.00 years
85

~f¥~W

WWW.GFEDU.NET

B. C. D.

1.94 years 1.87 years 1.76 years

Answer: B There are many types of duration used to measure a bond's price sensitivity to changes in interest rates. Choice B is the Macaulay duration. PCIF@8% PVofCF PVofCF as Weighted Year' CF % of Price Pved CF as % of Price 55.5556 1 60 0.9259 0.0576 0.057610242 2 1060 0.8573 908.7791 0.9424 1.884779516 Price: $964.33 Duration: 1.94 years

54. The option-adjusted duration of a callable bond will be close to the duration of a similar non-callable bond when the: A. Bond trades above the call price. B. Bond has a high volatility. C. Bond trades much lower than the call price. D. Bond trades above parity. Answer: C The option-adjusted spread is a basis point spread over the relevant Treasury yield less the value of the embedded call option. When interest rates are high, a callable bond is unlikely to be called This means that the value of the embedded call is close to zero and the callable bond duration is similar to the duration of the noncallable bond It also means that the bond is trading at a price lower than the call price. However, as rates fall, the value of the callable bend approaches the call price and the duration approaches zero. 55. When the maturity of a plain coupon bond increases, its duration increases: A. Indefinitely and regularly B. Up to a certain level C. Indefinitely and progressively D. In a way dependent on the bond being priced above or below par Answer: B A coupon bond's duration increases as its maturity increases, but it can never be larger than the number of years to maturity. 56. A portfolio has the following composition: I. Bond portfolio A: price $90,000, modified duration 2.5, long position in 8
86

ii£fj~1f

WWW.GFEDU.NET

bonds II. Bond portfolio B: price $110,000, modified duration 3, short position in 6 bonds III. Bond portfolio C: price $120,000, modified duration 3.3, long position in 12 bonds All interest rates are 10%. If the rates rise by 25 basis points, then the bond portfolio value will: A. Decrease by $11,43 B. Decrease by $21,330 C. Decrease by $12,573 D. Decrease by $23,463

Answer: A ~V(A) = -$90,000 x 8 x 2.5 x 0.0025 = -$4,500 (See Module 2, Topic 2.D)

~V(B) = -(-$110,000 x 6 x 3.0x 0.0025) = $4,950 ~V(C) = -$120,000 x 12 x 3.3 x 0.0025 = -$11,880
~V(portfoilo)

= -$4,500+ $4,950 - $11,800 = -$11,430

57. Of the following bonds, which one will suffer the largest proportional price increase after a 10 basis points decrease in interest rates? Assume the annual yield is 7%. A. A zero-coupon bond maturing in 5 years. B. A coupon paying bond, with Macaulay Duration of 3.81 years and Convexity of 16.39 years squared C. A bond with a coupon of 10% maturing in 10 years that is immediately callable. D. A Treasury Bill maturing in 6 months. Answer: A Tbe zero-coupon bond bas a Macauley duration of five years, which is tlie largest of the four securities, so its price will increase the most (in percentage terms) in response to a decrease in rates. 58. Consider the market environment in which I-year swaps are yielding 4% and have a duration of approximately 0.95. A I-year inverse floater with a coupon of 12 % - 3-month LIBOR has been just reset and is trading at par. The duration of this inverse floater will be CLOSEST to: A. 0.00. B. 1.90. C. 2.35. D. 2.85. Answer: C
87

~fj1_ttlf

WWW.GFEDU.NET

Every unit of this inverse floater is effectively composed from a long position on three par fixed coupon bonds (with a coupon rate of 5%) and a short position on two FRNs linked to LIB OR. The long position will have a negative duration of 0.95 x 3 == 2.85, while the short position will have a positive duration of 0.25 x 2 = 0.5. Therefore, the net duration of the inverse floater will be approximately 2.85 - 0.5 = 2.35. 59. Which of the following securities is likely to experience the greatest fall in price with a rise in interest rates (assuming that all of them have the same maturity and are currently priced to par)? A. Zero coupon bond B. Fixed coupon bond (7% coupon). C. Floating-rate note (3-month LlBOR) D. Inverse floater (14% - 3-month LIBOR). Answer: D The duration of the inverse floater will be almost twice as much as that of the fixed coupon bond, therefore its price will fall the most. The zero coupon bond will have the next highest duration, followed by the fixed coupon bond The floating-rate note will have the lowest (almost negligible) duration. 60. A trader buys a 7-year FRN, which pays interest rate linked to annual LIBOR. If the next LIBOR reset is one month away its duration will be closest to: A. 1 month. B. 1 year. C. 6 years. D. 7 years. Answer: A The duration of an FRN is close to the period remaining to the next LIBOR reset. 61. As the maturity of a bond rises, its price sensitivity: A. falls. B. nses. C. stays constant. D. rises or falls depending on the relative level of coupon. Answer: B As the maturity of the bond rises, its duration gets longer and therefore its price sensitivity increases. 62. As the coupon rate of a plain coupon bond is increased, its duration: A. decreases. B. increases up to a point and then decreases.
88

31~J.\E~1f WWW.GFEDU.NET

C. increases up to a point arid then stays constant. D. increases indefinitely. Answer: A As the coupon rate of a plain coupon bond is increased, its duration gradually falls to the level of an annuity (an annuity is effectively a coupon bond with an infinite coupon rate). 63. An 8-year 5 percent coupon bond with at par value of 100 is currently trading at a price of 94.65. The price of this bond rises to 96.35 when interest rates fall by 30 basis points and falls to 92.75 when interest rates rise by 30. The effective duration ofthis bond is CLOSEST to: A. 5.99 B. 6.34 C. 6.69 D. 7.04 Answer: B Effective duration
=

(Price if yield declines - Price if yield rises) (2 x Initial price x Change in yield)
=

(96.35 - 92.75) (2 x 94.65 x 0.003)

6.34.

64. Which of the following is NOT a criticism of duration as a risk measure? A. Duration is not an additive measure of risk. B. Duration is not a good measure of the risk of a MBS. C. Duration fails to account fQf differing volatilities associated with differing maturities, D. All of the above. Answer: A Duration is indeed an additive measure and this is a strong point in its favor for qualifying as a risk measure. The other two choices are the shortcomings of a duration measure. 65. Which of the following best describes the asset liability mix of a typical retail bank? A. Duration of assets is equal to duration of liabilities. B. Duration of assets is less than duration of liabilities. C. Asset and liability durations depend on interest rates. D. Duration of assets is more than duration of liabilities.

89

]t;f£~~

WWW.GFEDU.NET

Answer: D A large portion of the liabilities of a typical retail bank consist of demand deposits and other short term deposits. Whilst its assets consist of longer loans and bonds.

Modified Duration 66. Calculate the Modified Duration of a bond with a Macauley duration of 13.083 years. Assume market interest rates are 11.5% and the coupon on the bond is paid semi-annually. A. 13.083 B. 12.732 C. 12.459 D. 12.371 Answer: D 67. A 4.5-year 6% straight bond with annual coupon payments is traded at a clean price of 103.50. Calendar convention is 30E/360. The modified duration of the bond is 3.6%. The PVBP (Price value of a Basis Point) is: A. 0.0383 B. 0.0373 C. 0.0414 D. 0.0360 Answer: A Dirty price, dean price plus accrued interest = $103.50 + $3.00 - $106.50. PVBP

= D* P =

0.036 x$106.50 = 0.0383 WO

68. Which attribute of: a bond is NOT a reason for using effective duration instead of modified duration? A. Its life may be uncertain. B. Its cash flow may be uncertain. C. Its price volatility tends to decline as maturity approaches. D. It may include changes in adjustable rate coupons with caps or floors. Answer: C Price volatility does not affect the calculation of duration of a bond, while each of the factors in A, B, and D will affect duration. 69. Consider a 9 percent annual coupon 20-year bond trading at 6% with a price of 134.41. When rates rise 10bps, price reduces to 132.99 and when rates decrease by lObps, the price goes up to 135.85. What is the modified duration of
90

k~)(!_t&W

WWW.GFEDU.NET

the bond? A. 11.25. B. 10.61. C. 10.50. D. 10.73. Answer: B Modified duration estimates the relationship between the change in yields and the change in a price of a bond Choice B is closest to the correct answer. MD= (135.85-132.99) 2xO.00lx134.41 =10.64

70. Suppose the face value of a three-year option-free bond is USD 1,000 and the annual coupon is 10%. The current yield to maturity is 5%. What is the Modified Duration of this bond? A. 2.62 B. 2.85 C. 3.00 D. 2.75 Answer: A Given the annual coupon is 10% and the current yield to maturity is 5%, the price of the bond is given by: P=$100x(I/1.05)+$100 and the duration is: D=lx (100/1136.16)+2 x (100/1l36.16)+3x (110011136.16)=2.75.
=

x (111.052)+$1100

x (1/1.053)=$1136.16

net

The, modified duration is DJE! =t yield), or 2.7-5 i~] .(5)

2.62.

71. A bond is trading at a price of 100 with a yield of 8 percent. If the yield increases by 1 basis point, the price of the bond will decrease to 99.95. If the yield decreases by 1 basis point, the price of the bond will increase to 100.04. What is the modified duration of the bond? A. 5.0. B. -5.0 . . C. 4.5. D. -4.5. Answer: C With the information provided, two separate values for modified duration can be

91

~f¥~1f\VWW.GFEDU.NET

calculated

using the traditional

expression:

dPldv P

For a 1bp increase, the

modified duration is 5 = duration is 4= -

-0.05 . Similarly for the 1bp decrease, the modified 0.0001 x 100

0.040 . The effective duration for a bond may be determined -0.0001 x 100

as D fl.? = (V_- V+) , where V 0 is the original bond, V is the value of the bond after a

",

2Vody

~ y decrease in the yield, and V, is the value of the bond after a Ay increase in the

. yie Id H ere, we 1 lave:

100.04 - 99.95 = 4 .o. Thi IS the answer you Sh ou ld se Iect. - lIS . 1 2xlOOxO.000l

72. The Macaulay and modified duration of a 2-year annual pay bond paying an 8 percent coupon and priced to yield 10 percent are closest to: A. 1.92; 1.75. B. 1.75; 1.92. C. 8.00; 10.00. D. 1.08; 1.10. Answer: A N 1 2 Sum CF $8 $108 PVofCF $7.2727 $89.2562 $96.528
% of total PV

Nx % of total PV 0.0753 1.89494 1.9247;::;1.92

0.0753 0.9247 1.000

Modified duration = 1.92 = 1.75 1.10 73. AJ Treasury bond has a coupon rate of 6% per annum (the coupons are paid semi-annually) and a semi- annuaJly compounded yield of 4% per annum. Tile bond matures in 18 months and the next coupon will be paid 6 months from now. Which number below is closest to the bond's Macaulay duration? A. 1.023 years B. 1.457 years C. 1.500 years D. 2.915 years Answer: B Period
1

2 3 Total
92

Cash Flows $3.000 $3.000 $103.000

PVCF@2% $2.9412 $2.8835 $97.0592 $102.8839

txPVCF 2.9412 5.7670 291.1776 299.8858

~:m~1'f WWW.GFEDU.NET Modified duration = 1 x _1_ x 299.8858 = 2.8576 periods = 1.4288 years 102.8839 1.02 Macauley duration == 1.02 x 1.4288 = 1.4574 years

74. A zero-coupon bond with a maturity of 10 years has an annual effective yield of 10%. What is the closest value for its modified duration? A.9 B.I0 C.100 D. Insufficient Information Answer: A You must first recall that the Macauley duration of a zero-coupon bond is equal to its maturity. Then, the modified duration of a zero-coupon bond is: Macauley duration = JQ_ = 9.09 l+i 1.10 1+ i B. Incorrect. It corresponds to the Macauley duration, not the Modified duration. C. Incorrect. The denominator used in the formula was i instead of 1+i. D. Incorrect. All the necessary information is in there. A. Correct. The above formula was used correctly, Dmod = Macauley duration

Convexity 75. A convexity adjustment A. bond prices, B sVap rates. C. Eurodollar futures. D. currency futures. is necessary to compute yields from:

.n t

Answer: C A convexity adjustment is necessary to convert Eurodollar forward rates into forward interest rates. 76. Coupon curve duration is a useful method to estima~e convexity from market prices of an MBS. Assume the coupon curve of prices for Ginnie Maes in June 2001 is as follows: 6 percent at 92, 7 percent at 94, and 8 percent at 96.5. What is the estimated convexity of the 7s? A. 53. B. 26. C. 13.
93

Das könnte Ihnen auch gefallen