Академический Документы
Профессиональный Документы
Культура Документы
(Chapter 6)
Markowitz Model
Employment of Factor Models
Essence of the Single-Factor Model
The Characteristic Line
Expected Return in the Single-Factor Model
Single-Factor Model’s Simplified Formula for
Portfolio Variance
Explained Versus Unexplained Variance
Multi-Factor Models
Models for Estimating Expected Return
Markowitz Model
M M
σ 2 (rp ) x x Cov(r , r )
j1 k 1
j k j k
σ 2 (r ) Cov(r1 , r2 ) . . . Cov(r1 , rM )
1
Cov(r , r ) σ 2 (r2 ) . . . Cov(r2 , rM )
2 1
. . . .
. . . .
2
Cov(r M , r1 ) . . . . . . σ (rM )
Problem: Tremendous data requirement.
Number of security variances needed = M.
Number of covariances needed = (M2 - M)/2
Total = M + (M2 - M)/2
Example: (100 securities)
100 + (10,000 - 100)/2 = 5,050
Therefore, in order for modern portfolio theory to be usable for
large numbers of securities, the process had to be simplified.
(Years ago, computing capabilities were minimal)
Employment of Factor Models
To generate the efficient set, we need estimates of
expected return and the covariances between the
securities in the available population. Factor models
may be used in this regard.
Risk Factors – (rate of inflation, growth in industrial
production, and other variables that induce stock prices
to go up and down.)
May be used to evaluate covariances of return between
securities.
Expected Return Factors – (firm size, liquidity, etc.)
May be used to evaluate expected returns of the securities.
In the discussion that follows, we first focus on risk
factor models. Then the discussion shifts to factors
affecting expected security returns.
Essence of the Single-Factor
Model
Fluctuations in the return of a security relative to that of
another (i.e., the correlation between the two) do not depend
upon the individual characteristics of the two securities.
Instead, relationships (covariances) between securities occur
because of their individual relationships with the overall
market (i.e., covariance with the market).
If Stock (A) is positively correlated with the market, and if
Stock (B) is positively correlated with the market, then Stocks
(A) and (B) will be positively correlated with each other.
Given the assumption that covariances between securities
can be accounted for by the pull of a single common factor
(the market), the covariance between any two stocks can be
written as:
Cov(rj , rk ) β jβk σ 2 (rM )
The Characteristic Line
(See Chapter 3 for a Review of the Statistics)
rj,t
0.4
0.3
0.2 = j
0.1
Aj
0 rM,t
0 0.2 0.4 0.6
The Characteristic Line
(Continued)
Note: A stock’s return can be broken down into two
parts:
Movement along the characteristic line (changes
in the stock’s returns caused by changes in the
market’s returns).
Deviations from the characteristic line (changes in
the stock’s returns caused by events unique to the
individual stock).
rj,t A j β jrM,t ε j,t
-10
Expected Return in the Single-Factor
Model
Actual Returns: rj,t A j β jrM, t ε j,t
Expected Residual:
Given the characteristic line is truly the line of best
fit, the sum of the residuals
n
would be equal to zero
ε
t 1
j,t 0
n
=
i =1
h i (β j[rM, i E(rM )] ε j,i ) 2
n
=
i =1
h i (β 2j [rM, i E(rM )]2 2 β j[rM, i E(rM )] ε j,i ε 2j,i )
n n n
= β 2j
i =1
h i [rM, i E(rM )]2 2 β j
i 1
h i [rM,i E(rM )] ε j,i
i 1
h i ε 2j,i
Note: n
β 2j
i =1
h i [rM,i E(rM )]2 β 2jσ 2 (rM )
n n
2βj h [r
i 1
i M,i E(rM )] ε j,i 2 β j h [r
i 1
i M, i E(rM )][ε j,i E(ε j )]
Since E(ε j ) 0
= 2 β jCov(rM , ε j )
=0
Since Cov(rM , ε j ) is equal to zero for the
best fit line in simple regression .
n
i 1
h i ε 2j,i σ 2 (ε j ) Residual Variance (See Chapter 3)
Therefore:
σ 2 (rj ) β 2jσ 2 (rM ) σ 2 (ε j )
Variance of a Portfolio
Same equation as the one for individual security
variance: σ 2 (r ) β 2 σ 2 (r ) σ 2 (ε )
p p M p
1
8
0.8
6
0.6
4
0.4
2
0.2
0 0
1 5 9 13 17 1 5 9 13 17
Number of Stocks Number of Stocks
Explained Vs. Unexplained Variance
(A Two Stock Portfolio Example)
β 2 σ 2 (r ) β Aβ Bσ 2 (rM ) σ 2 (ε ) Cov(ε A , ε B )
A M A
β β σ 2 (r ) βB σ (rM )
2 2 Cov(ε , ε ) σ 2
(ε )
B A M B A B
Covariance Matrix for Covariance Matrix for
Explained Variance Unexplained Variance
Explained Unexplained
2
m m
=
j=1
x jβ j σ 2 (rM )
j =1
x 2j σ 2 (ε j )
= β p2 σ 2 (rM ) σ 2 (ε p )
If Cov(ε A , ε B ) 0, residual variance will be understated .
If Cov(ε A , ε B ) 0, residual variance will be overstated .
A Note on Residual Variance
The Single-Factor Model assumes zero correlation
between residuals:
Cov(ε j , ε k ) 0
In this case, portfolio residual variance is expressed
as: m
2 2 2
σ (ε p ) x σ (ε )
j1
j j
σ 2 (rp ) β g,
2
p σ 2
(rg ) β 2
I,p σ 2
(rI ) σ 2
(ε p )
m
where: β g,p x β
j1
j g, j
m
β I,p x β
j1
j I, j
m
σ 2 (ε p )
j1
x 2j σ 2 (ε j )
Assuming Cov(ε j , ε k ) 0
2 (rp ) g
2 2 (r ) 2 2 (r ) 2
,p g I, p I g, p I, p Cov (rg , rI )
E(rj ) A j β jE(rM )
Given a firm’s estimated characteristic line and an
estimate of the future return on the market, the
security’s expected return can be calculated.
Models for Estimating Expected Return
(Continued)
Technical Analysis
Analyze past patterns of market data (e.g., price
changes) in order to predict future patterns of
market data. “Volumes have been written on this
subject.
Size Effect
Returns on small stocks (small market value) tend
to be superior to returns on large stocks. Note:
Small NYSE stocks tend to outperform small
NASDAQ stocks.
January Effect
Abnormally high returns tend to be earned
(especially on small stocks) during the month of
January.
Other Factors That May Be Used in Predicting Expected
Return (continued)
Future Estimate: