Sie sind auf Seite 1von 42

CMU-logo

Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Lecture 2.6: Implementing the Portfolio
Problem
Investment Analysis
Fall, 2012
Anisha Ghosh
Tepper School of Business
Carnegie Mellon University
November 8, 2012
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Readings and Assignments
Chapters 7 and 8 of the course textbook (EGBG) cover the
material I will be discussing in this lecture.
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Outline
Inputs to Portfolio Analysis
Models for Forecasting Correlation Structures
1
Single-Index Models
2
Multi-Index Models
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Inputs to Portfolio Analysis
To dene the efcient frontier, we need to compute the expected return
and standard deviation of return on a portfolio.
The expected return and standard deviation of a portfolio is given by:
E (R
p
) =
N

i =1
X
i
E (R
i
)

p
=
_
_
_
_
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

ij
_

_
1/2
the input data necessary to perform portfolio analysis includes
estimates of:
Expected return on each security
Variance of each security
The correlation between each possible pairs of securities
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Inputs to Portfolio Analysis
To dene the efcient frontier, we need to compute the expected return
and standard deviation of return on a portfolio.
The expected return and standard deviation of a portfolio is given by:
E (R
p
) =
N

i =1
X
i
E (R
i
)

p
=
_
_
_
_
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

ij
_

_
1/2
the input data necessary to perform portfolio analysis includes
estimates of:
Expected return on each security
Variance of each security
The correlation between each possible pairs of securities
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Inputs to Portfolio Analysis
To dene the efcient frontier, we need to compute the expected return
and standard deviation of return on a portfolio.
The expected return and standard deviation of a portfolio is given by:
E (R
p
) =
N

i =1
X
i
E (R
i
)

p
=
_
_
_
_
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

ij
_

_
1/2
the input data necessary to perform portfolio analysis includes
estimates of:
Expected return on each security
Variance of each security
The correlation between each possible pairs of securities
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Inputs to Portfolio Analysis
To dene the efcient frontier, we need to compute the expected return
and standard deviation of return on a portfolio.
The expected return and standard deviation of a portfolio is given by:
E (R
p
) =
N

i =1
X
i
E (R
i
)

p
=
_
_
_
_
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

ij
_

_
1/2
the input data necessary to perform portfolio analysis includes
estimates of:
Expected return on each security
Variance of each security
The correlation between each possible pairs of securities
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Problems with Estimating Correlations
Let N denote the number stocks to be included in the portfolio.
the number of correlations to be estimated is
N(N1)
2
Most nancial institutions follow between 150 and 250 stocks they
need to estimate between 11, 175 and 31, 125 correlations!!
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Problems with Estimating Correlations
Let N denote the number stocks to be included in the portfolio.
the number of correlations to be estimated is
N(N1)
2
Most nancial institutions follow between 150 and 250 stocks they
need to estimate between 11, 175 and 31, 125 correlations!!
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Problems with Estimating Correlations
Let N denote the number stocks to be included in the portfolio.
the number of correlations to be estimated is
N(N1)
2
Most nancial institutions follow between 150 and 250 stocks they
need to estimate between 11, 175 and 31, 125 correlations!!
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Single-Index Models
Intuition: Security returns are correlated because of a common
response to market changes, and a measure of this correlation can be
obtained by relating the return on a stock to the return on a stock
market index
Basic Equation:
R
i
=
i
+
i
R
m
+
i
, i = 1, 2, ..., N

i
+
i
is the component of security i s return that is independent
of the markets performance
R
m
is the rate of return on the market index

i
is a constant that measures the expected change in R
i
given a
change in R
m
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Single-Index Models
Intuition: Security returns are correlated because of a common
response to market changes, and a measure of this correlation can be
obtained by relating the return on a stock to the return on a stock
market index
Basic Equation:
R
i
=
i
+
i
R
m
+
i
, i = 1, 2, ..., N

i
+
i
is the component of security i s return that is independent
of the markets performance
R
m
is the rate of return on the market index

i
is a constant that measures the expected change in R
i
given a
change in R
m
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Single-Index Models contd.
By Construction:
E (
i
) = 0, i = 1, 2, ..., N
By Assumption:
1
Index unrelated to unique return:
E [
i
(R
m
E (R
m
))] = 0, i = 1, 2, ..., N
2
Securities only related through common response to market:
E (
i

j
) = 0, i , j = 1, 2, ..., N
i =j
By Denition:
1
Variance of
i
:
E
_

2
i
_
=
2
i
, i = 1, 2, ..., N
2
Variance of R
m
:
E
_
(R
m
E (R
m
))
2
_
=
2
m
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Single-Index Models contd.
By Construction:
E (
i
) = 0, i = 1, 2, ..., N
By Assumption:
1
Index unrelated to unique return:
E [
i
(R
m
E (R
m
))] = 0, i = 1, 2, ..., N
2
Securities only related through common response to market:
E (
i

j
) = 0, i , j = 1, 2, ..., N
i =j
By Denition:
1
Variance of
i
:
E
_

2
i
_
=
2
i
, i = 1, 2, ..., N
2
Variance of R
m
:
E
_
(R
m
E (R
m
))
2
_
=
2
m
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Single-Index Models contd.
By Construction:
E (
i
) = 0, i = 1, 2, ..., N
By Assumption:
1
Index unrelated to unique return:
E [
i
(R
m
E (R
m
))] = 0, i = 1, 2, ..., N
2
Securities only related through common response to market:
E (
i

j
) = 0, i , j = 1, 2, ..., N
i =j
By Denition:
1
Variance of
i
:
E
_

2
i
_
=
2
i
, i = 1, 2, ..., N
2
Variance of R
m
:
E
_
(R
m
E (R
m
))
2
_
=
2
m
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Expected Return on a Security
The expected return on a security is
E (R
i
) = E (
i
+
i
R
m
+
i
)
= E (
i
) + E (
i
R
m
) + E (
i
)
=
i
+
i
E (R
m
)
the expected return has two components: a unique part
i
and a
market-related part
i
E (R
m
)
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Expected Return on a Security
The expected return on a security is
E (R
i
) = E (
i
+
i
R
m
+
i
)
= E (
i
) + E (
i
R
m
) + E (
i
)
=
i
+
i
E (R
m
)
the expected return has two components: a unique part
i
and a
market-related part
i
E (R
m
)
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Variance of a Security
The variance of a security is

2
i
= E
_
(R
i
E (R
i
))
2
_
= E [(
i
+
i
R
m
+
i
) (
i
+
i
E (R
m
))]
2
= E [
i
(R
m
E (R
m
)) +
i
]
2
=
2
i
E
_
(R
m
E (R
m
))
2
_
+ E [
i
]
2
+ 2
i
E [
i
(R
m
E (R
m
))]
=
2
i
E
_
(R
m
E (R
m
))
2
_
+ E [
i
]
2
=
2
i

2
m
+
2
i
a securitys variance has two parts: unique risk and market-related
risk
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Variance of a Security
The variance of a security is

2
i
= E
_
(R
i
E (R
i
))
2
_
= E [(
i
+
i
R
m
+
i
) (
i
+
i
E (R
m
))]
2
= E [
i
(R
m
E (R
m
)) +
i
]
2
=
2
i
E
_
(R
m
E (R
m
))
2
_
+ E [
i
]
2
+ 2
i
E [
i
(R
m
E (R
m
))]
=
2
i
E
_
(R
m
E (R
m
))
2
_
+ E [
i
]
2
=
2
i

2
m
+
2
i
a securitys variance has two parts: unique risk and market-related
risk
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Covariance Between Securities
The covariance between securities i and j is

ij
= E
_
[(
i
+
i
R
m
+
i
) (
i
+
i
E (R
m
))]
[(
j
+
j
R
m
+
j
) (
j
+
j
E (R
m
))]
_
= E {[
i
(R
m
E (R
m
)) +
i
] [
j
(R
m
E (R
m
)) +
j
]}
=
i

j
E
_
(R
m
E (R
m
))
2
_
+
i
E [
j
(R
m
E (R
m
))]
+
j
E [
i
(R
m
E (R
m
))] + E (
i

j
)
=
i

2
m
the covariance depends only on market risk.
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Covariance Between Securities
The covariance between securities i and j is

ij
= E
_
[(
i
+
i
R
m
+
i
) (
i
+
i
E (R
m
))]
[(
j
+
j
R
m
+
j
) (
j
+
j
E (R
m
))]
_
= E {[
i
(R
m
E (R
m
)) +
i
] [
j
(R
m
E (R
m
)) +
j
]}
=
i

j
E
_
(R
m
E (R
m
))
2
_
+
i
E [
j
(R
m
E (R
m
))]
+
j
E [
i
(R
m
E (R
m
))] + E (
i

j
)
=
i

2
m
the covariance depends only on market risk.
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Portfolios
The expected return on any portfolio if the single-index model holds is
E (R
p
) =
N

i =1
X
i
E (R
i
)
=
N

i =1
X
i

i
+
N

i =1
X
i

i
E (R
m
)
The variance of a portfolio of stocks is given by

2
p
=
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

ij
=
N

i =1
X
2
i

2
i

2
m
+
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

2
m
estimation of the expected return and risk of any portfolio requires a
total of 3N + 2 estimates (the estimates of the
i
,
i
, and
2
i
for each
stock, and an estimate of the expected return and variance of the
market)
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Portfolios
The expected return on any portfolio if the single-index model holds is
E (R
p
) =
N

i =1
X
i
E (R
i
)
=
N

i =1
X
i

i
+
N

i =1
X
i

i
E (R
m
)
The variance of a portfolio of stocks is given by

2
p
=
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

ij
=
N

i =1
X
2
i

2
i

2
m
+
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

2
m
estimation of the expected return and risk of any portfolio requires a
total of 3N + 2 estimates (the estimates of the
i
,
i
, and
2
i
for each
stock, and an estimate of the expected return and variance of the
market)
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Portfolios
The expected return on any portfolio if the single-index model holds is
E (R
p
) =
N

i =1
X
i
E (R
i
)
=
N

i =1
X
i

i
+
N

i =1
X
i

i
E (R
m
)
The variance of a portfolio of stocks is given by

2
p
=
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

ij
=
N

i =1
X
2
i

2
i

2
m
+
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

2
m
estimation of the expected return and risk of any portfolio requires a
total of 3N + 2 estimates (the estimates of the
i
,
i
, and
2
i
for each
stock, and an estimate of the expected return and variance of the
market)
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Characteristics of the Single-Index Model
The return of any portfolio if the single-index model holds is
E (R
p
) =
p
+
p
E (R
m
)
The Beta on the portfolio is a weighted-average of the individual
i
s on
each stock in the portfolio where the weights are the fraction of the
portfolio invested in each stock:

p
=
N

i =1
X
i

i
The Alpha on the portfolio is similarly dened:

p
=
N

i =1
X
i

i
the Beta on the market is 1 and stocks are thought of as being more
or less risky than the market, according to whether their Beta is larger
or smaller than 1
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Characteristics of the Single-Index Model
The return of any portfolio if the single-index model holds is
E (R
p
) =
p
+
p
E (R
m
)
The Beta on the portfolio is a weighted-average of the individual
i
s on
each stock in the portfolio where the weights are the fraction of the
portfolio invested in each stock:

p
=
N

i =1
X
i

i
The Alpha on the portfolio is similarly dened:

p
=
N

i =1
X
i

i
the Beta on the market is 1 and stocks are thought of as being more
or less risky than the market, according to whether their Beta is larger
or smaller than 1
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Characteristics of the Single-Index Model
The return of any portfolio if the single-index model holds is
E (R
p
) =
p
+
p
E (R
m
)
The Beta on the portfolio is a weighted-average of the individual
i
s on
each stock in the portfolio where the weights are the fraction of the
portfolio invested in each stock:

p
=
N

i =1
X
i

i
The Alpha on the portfolio is similarly dened:

p
=
N

i =1
X
i

i
the Beta on the market is 1 and stocks are thought of as being more
or less risky than the market, according to whether their Beta is larger
or smaller than 1
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Characteristics of the Single-Index Model
The return of any portfolio if the single-index model holds is
E (R
p
) =
p
+
p
E (R
m
)
The Beta on the portfolio is a weighted-average of the individual
i
s on
each stock in the portfolio where the weights are the fraction of the
portfolio invested in each stock:

p
=
N

i =1
X
i

i
The Alpha on the portfolio is similarly dened:

p
=
N

i =1
X
i

i
the Beta on the market is 1 and stocks are thought of as being more
or less risky than the market, according to whether their Beta is larger
or smaller than 1
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Diversiable vs Nondiversiable Risk
The variance on the portfolio can be written as

2
p
=
N

i =1
X
2
i

2
i

2
m
+
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

2
m
=
N

i =1
N

j =1
X
i
X
j

2
m
+
N

i =1
X
2
i

2
i
=
_
_
_
N

i =1
X
i

i
_
_
_
_
_
_
_
N

j =1
X
j

j
_
_
_
_

2
m
+
N

i =1
X
2
i

2
i
=
2
p

2
m
+
N

i =1
X
2
i

2
i
If equal amounts on money are placed in the N stocks,

2
p
=
2
p

2
m
+
1
N
N

i =1
1
N

2
i
. .
0 as N
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Diversiable vs Nondiversiable Risk
The variance on the portfolio can be written as

2
p
=
N

i =1
X
2
i

2
i

2
m
+
N

i =1
X
2
i

2
i
+
N

i =1
N

j =1
i =j
X
i
X
j

2
m
=
N

i =1
N

j =1
X
i
X
j

2
m
+
N

i =1
X
2
i

2
i
=
_
_
_
N

i =1
X
i

i
_
_
_
_
_
_
_
N

j =1
X
j

j
_
_
_
_

2
m
+
N

i =1
X
2
i

2
i
=
2
p

2
m
+
N

i =1
X
2
i

2
i
If equal amounts on money are placed in the N stocks,

2
p
=
2
p

2
m
+
1
N
N

i =1
1
N

2
i
. .
0 as N
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Diversiable vs Nondiversiable Risk contd.
as we hold larger and larger portfolios, the risk of the portfolio
approaches

p
=
p

2
i
is referred to as diversiable risk while
i
is a measure of a
securitys nondiversiable risk or systematic risk
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Diversiable vs Nondiversiable Risk contd.
as we hold larger and larger portfolios, the risk of the portfolio
approaches

p
=
p

2
i
is referred to as diversiable risk while
i
is a measure of a
securitys nondiversiable risk or systematic risk
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Estimating Historical Betas
The
i
and
i
of a rm can be estimated using regression analysis
using historical returns on the security and the market:

i
=

im

2
m
=
T

t =1
_
R
it

1
T
T

t =1
R
it
__
R
mt

1
T
T

t =1
R
mt
_
T

t =1
_
R
mt

1
T
T

t =1
R
mt
_
2

i
=
_
1
T
T

t =1
R
it
_

i
_
1
T
T

t =1
R
mt
_
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Multi-Index Models
Intuition: Stock prices move together not only because of common
movement with the market but also because of inuences beyond the
market.
Basic Equation:
R
i
= a
i
+ b
i 1
I
1
+ b
i 2
I
2
+ b
i 3
I
3
+ ... + b
iL
I
L
+ c
i
, i = 1, 2, ..., N
By Denition:
1
Residual variance of stock i :
E
_
c
2
i
_
=
2
ci
, i = 1, 2, ..., N
2
Variance of index j :
E
_
(I
j
E (I
j
))
2
_
=
2
Ij
, j = 1, 2, ..., L
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Multi-Index Models
Intuition: Stock prices move together not only because of common
movement with the market but also because of inuences beyond the
market.
Basic Equation:
R
i
= a
i
+ b
i 1
I
1
+ b
i 2
I
2
+ b
i 3
I
3
+ ... + b
iL
I
L
+ c
i
, i = 1, 2, ..., N
By Denition:
1
Residual variance of stock i :
E
_
c
2
i
_
=
2
ci
, i = 1, 2, ..., N
2
Variance of index j :
E
_
(I
j
E (I
j
))
2
_
=
2
Ij
, j = 1, 2, ..., L
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Multi-Index Models
Intuition: Stock prices move together not only because of common
movement with the market but also because of inuences beyond the
market.
Basic Equation:
R
i
= a
i
+ b
i 1
I
1
+ b
i 2
I
2
+ b
i 3
I
3
+ ... + b
iL
I
L
+ c
i
, i = 1, 2, ..., N
By Denition:
1
Residual variance of stock i :
E
_
c
2
i
_
=
2
ci
, i = 1, 2, ..., N
2
Variance of index j :
E
_
(I
j
E (I
j
))
2
_
=
2
Ij
, j = 1, 2, ..., L
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Multi-Index Models contd.
By Construction:
1
Mean of c
i
:
E (c
i
) = 0, i = 1, 2, ..., N
2
Covariances between indexes j and k:
E [(I
j
E (I
j
)) (I
k
E (I
k
))] = 0, j , k = 1, 2, ..., N
j =k
3
Covariance between the residual for stock i and index j :
E [c
i
(I
j
E (I
j
))] = 0, i = 1, 2, ..., N
j =1,2,...,L
By Assumption:
1
Covariance between c
i
and c
j
:
E (c
i
c
j
) = 0, i , j = 1, 2, ..., N
i =j
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Multi-Index Models contd.
By Construction:
1
Mean of c
i
:
E (c
i
) = 0, i = 1, 2, ..., N
2
Covariances between indexes j and k:
E [(I
j
E (I
j
)) (I
k
E (I
k
))] = 0, j , k = 1, 2, ..., N
j =k
3
Covariance between the residual for stock i and index j :
E [c
i
(I
j
E (I
j
))] = 0, i = 1, 2, ..., N
j =1,2,...,L
By Assumption:
1
Covariance between c
i
and c
j
:
E (c
i
c
j
) = 0, i , j = 1, 2, ..., N
i =j
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Characteristics of Securities
When the multi-index model describes the return structure, the
expected return of a security is:
E (R
i
) = a
i
+ b
i 1
E (I
1
) + b
i 2
E (I
2
) + ... + b
iL
E (I
L
)
The variance of return is:

2
i
= b
2
i 1

2
I1
+ b
2
i 2

2
I2
+ ... + b
2
iL

2
IL
+
2
ci
Covariance between security i and j is:

ij
= b
i 1
b
j 1

2
I1
+ b
i 2
b
j 2

2
I2
+ ... + b
iL
b
jL

2
IL
the number of inputs required is 2N + 2L + LN (the expected return
and variance of each stocks return, each index loading on the L factors
or indexes, and the means and variances of each index)
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Characteristics of Securities
When the multi-index model describes the return structure, the
expected return of a security is:
E (R
i
) = a
i
+ b
i 1
E (I
1
) + b
i 2
E (I
2
) + ... + b
iL
E (I
L
)
The variance of return is:

2
i
= b
2
i 1

2
I1
+ b
2
i 2

2
I2
+ ... + b
2
iL

2
IL
+
2
ci
Covariance between security i and j is:

ij
= b
i 1
b
j 1

2
I1
+ b
i 2
b
j 2

2
I2
+ ... + b
iL
b
jL

2
IL
the number of inputs required is 2N + 2L + LN (the expected return
and variance of each stocks return, each index loading on the L factors
or indexes, and the means and variances of each index)
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Characteristics of Securities
When the multi-index model describes the return structure, the
expected return of a security is:
E (R
i
) = a
i
+ b
i 1
E (I
1
) + b
i 2
E (I
2
) + ... + b
iL
E (I
L
)
The variance of return is:

2
i
= b
2
i 1

2
I1
+ b
2
i 2

2
I2
+ ... + b
2
iL

2
IL
+
2
ci
Covariance between security i and j is:

ij
= b
i 1
b
j 1

2
I1
+ b
i 2
b
j 2

2
I2
+ ... + b
iL
b
jL

2
IL
the number of inputs required is 2N + 2L + LN (the expected return
and variance of each stocks return, each index loading on the L factors
or indexes, and the means and variances of each index)
CMU-logo
Today Inputs to Portfolio Analysis Single Index Models Multi Index Models
Characteristics of Securities
When the multi-index model describes the return structure, the
expected return of a security is:
E (R
i
) = a
i
+ b
i 1
E (I
1
) + b
i 2
E (I
2
) + ... + b
iL
E (I
L
)
The variance of return is:

2
i
= b
2
i 1

2
I1
+ b
2
i 2

2
I2
+ ... + b
2
iL

2
IL
+
2
ci
Covariance between security i and j is:

ij
= b
i 1
b
j 1

2
I1
+ b
i 2
b
j 2

2
I2
+ ... + b
iL
b
jL

2
IL
the number of inputs required is 2N + 2L + LN (the expected return
and variance of each stocks return, each index loading on the L factors
or indexes, and the means and variances of each index)

Das könnte Ihnen auch gefallen