Asset Pricing Models

The Capital Asset Pricing Model

 The capital asset pricing model (CAPM) extends

capital market theory in a way that allows investors to
evaluate the risk–return trade-off for both diversified
portfolios and individual securities
• The CAPM:
• Redefines the relevant measure of risk from total volatility to just
the nondiversifiable portion of that total volatility (systematic risk)
• The risk measure is called the beta coefficient and calculates the
level of a security’s systematic risk compared to that of the market
portfolio

1
 A Conceptual Development of the
CAPM
 The existence of a risk-free asset resulted in deriving a
capital market line (CML) that became the relevant
frontier
 However, CML cannot be used to measure the expected
return on an individual asset
 For individual asset (or any portfolio), the relevant risk
measure is the asset’s covariance with the market
portfolio
 That is, for an individual asset i, the relevant risk is not σi,
but rather σi riM, where riM is the correlation coefficient
between the asset and the market

2
 A Conceptual Development of the
CAPM
 Inserting this product into the CML and adapting the
notation for the ith individual asset:
 E ( RM ) − RFR 
E ( Ri ) = RFR + ( i riM )  
 M 
• Let βi=(σi riM) / σM be the asset beta measuring the relative risk
with the market, the systematic risk
E ( Ri ) = RFR + i  E ( RM ) − RFR 

• The CAPM indicates what should be the expected or required

rates of return on risky assets
• This helps to value an asset by providing an appropriate discount
rate to use in dividend valuation models

3
 The Security Market Line

 The SML
• Is a graphical form of the CAPM
• Shows the trade-off between risk and expected return as a
straight line intersecting the vertical axis (zero-risk point) at
the risk-free rate
• Considers only the systematic component of an
investment’s volatility
• Can be applied to any individual asset or collection of
assets

4
The Security Market Line

5
 The Security Market Line

 Identifying Undervalued and Overvalued Assets

• In equilibrium, all assets and all portfolios of assets should
plot on the SML
• Any security with an estimated return that plots above the
SML is underpriced
• Any security with an estimated return that plots below the
SML is overpriced
• A superior investor must derive value estimates for assets
that are consistently superior to the consensus market
evaluation to earn better risk-adjusted rates of return than
the average investor

6
The Security Market Line

7
The Security Market Line

8
 The Security Market Line

 Calculating Systematic Risk

• A beta coefficient for Security i can be calculated directly
from the following formula:
𝜎𝑖 𝐶𝑂𝑉(𝑅𝑖 , 𝑅𝑀 )
𝛽𝑖 = 𝛾𝑖𝑀 = 2
𝜎𝑀 𝜎𝑀

• Security betas can also be estimated as the slope coefficient

in a regression equation between the returns to the security
(Rit) over time and the returns (RMt) to the market portfolio
(the security’s characteristic line):
Rit = ai + i ( RMt ) +  it

9
 The Security Market Line

 Computing a Characteristic Line: An Example

• The example shows how to estimate a characteristic line for
Microsoft Corp (MSFT) using monthly return data from
January 2016 to December 2016
• Betas for MSFT are calculated using:
• The S&P 500 (SPX)
• The MSCI World Equity (MXWO) index

10
The Security Market Line

11
The Security Market Line

12
The Security Market Line

13
 The Security Market Line

 Industry Characteristic Lines

• The characteristic line used to estimate beta value can be
computed for sector indexes

14
 Empirical Tests of the CAPM

 Stability of beta
• Numerous studies have examined the stability of beta and
generally concluded that the risk measure was not stable for
individual stocks but was stable for portfolios of stocks
• The larger the portfolio and the longer the period, the
more stable the beta estimate
• The betas tended to regress toward the mean
• High-beta portfolios tended to decline over time toward
1.00, whereas low beta portfolios tended to increase
over time toward unity

15
 Empirical Tests of the CAPM
 Relationship between the systematic risk and return
• Study (Jensen, 1972) shows that:
• Most of the measured SMLs had a positive slope
• The slopes change between periods
• The intercepts are not zero
• The intercepts change between periods

16
Empirical Tests of the CAPM

17
 Arbitrage Pricing Theory

 CAPM is criticized because of

• The many unrealistic assumptions
• The difficulties in selecting a proxy for the market portfolio
as a benchmark
 An alternative pricing theory with fewer assumptions
was developed: Arbitrage Pricing Theory (APT)

18
 Arbitrage Pricing Theory

 Major assumptions:
• Capital markets are perfectly competitive
• Investors always prefer more wealth to less wealth with
certainty
• The stochastic process generating asset returns can be
expressed as a linear function of a set of K factors or
indexes
 In contrast to CAPM, APT does not assume:
• Normally distributed security returns
• Quadratic utility function
• A mean-variance efficient market portfolio

19
 Arbitrage Pricing Theory
 Theory assumes that the return-generating process
can be represented as a K factor model of the form:

Ri = E ( Ri ) + bi11 + bi 2 2 + + bik  k +  i for i = 1 to n

where:
Ri = actual return on Asset i during a specified time period, i = 1, 2, 3, … n
E(Ri) = expected return for Asset i if all the risk factors have zero changes
bij = reaction in Asset i’s returns to movements in a common risk factor j
δk = set of common factors or indexes with a zero mean that influences the
returns on all assets
εi = unique effect on Asset i’s return (a random error term that, by assumption,
is completely diversifiable in large portfolios and has a mean of zero)
n = number of assets

20
 Arbitrage Pricing Theory
 The APT requires that in equilibrium the return on a zero-
investment, zero-systematic-risk portfolio is zero when
the unique effects are fully diversified
 This assumption implies that the expected return on any
Asset i can be expressed as:

E ( Ri ) =  0 + 1bi1 +  2bi 2 + +  k bik ( APT)

where:
λ0 = expected return on an asset with zero systematic risk
λk = risk premium related to the jth common risk factor
bik = pricing relationship between the risk premium and the asset;
that is, how responsive Asset i is to the jth common factor.
(These are called factor betas or factor loadings.)

21
Arbitrage Pricing Theory

22
 Arbitrage Pricing Theory
 Two-stock and a two-factor model example:
• Assume that there are two common factors: one related to
unexpected changes in the level of inflation and another
related to unanticipated changes in the real level of GDP
• Risk factor definitions and sensitivities:
• δ1 = unanticipated changes in the rate of inflation. The risk premium
related to this factor is 2 percent for every 1 percent change in the rate
(λ1 = 0.02).
• δ2 = unexpected changes in the growth rate of real GDP. The average
risk premium related to this factor is 3 percent for every 1 percent
change in the rate growth (λ2 = 0.03).
• λ0 = rate of return on a zero-systematic risk asset (zero-beta) is 4
percent (λ0 = 0.04).

23
Arbitrage Pricing Theory

24
 Security Valuation with the APT
 Example
• Suppose that three stocks (A, B, and C) and two common
systematic risk factors (1 and 2) have the following relationship
(for simplicity, it is assumed that the zero-beta return [λ0] equals
zero):
E(RA) = (0.80)λ1 + (0.90)λ2
E(RB) = (−0.20)λ1 + (1.30)λ2
E(RC) = (1.80)λ1 + (0.50)λ2

• If λ1 = 4 percent and λ2 = 5 percent, then the returns expected by

the market over the next year can be expressed as:
E(RA) = (0.80)(4%) + (0.90)(5%) = 7.7%
E(RB) = (−0.20)(4%) + (1.30)(5%) = 5.7%
E(RC) = (1.80)(4%) + (0.50)(5%) = 9.7%

25
 Multifactor Models

 In a multifactor model, the investor chooses the exact

number and identity of risk factors, while the APT
model does not specify either of them
Rit = ai +  bi1 F1t + bi 2 F2t + + biK FKt  + eit
where:
Fit = Period t return to the jth designated risk factor
Rit = Security i’s return that can be measured as either a nominal
or excess return to Security i

26
 Multifactor Models in Practice

 A wide variety of empirical factor specifications have

been employed in practice
 Alternative models attempt to identify a set of
economic influences
 Two approaches:
• Risk factors can be viewed as macroeconomic in nature
• Risk factors can also be viewed at a microeconomic level

27
 Multifactor Models in Practice

 Macroeconomic-Based Risk Factor Models

• Chen, Roll, and Ross (1986):
Rit = ai +  bi1 RMt + bi 2 MPt + bi 3 DEI t + bi 4UI t + bi 5UPRt + bi 6UTSt  + eit

Where:
RM = return on a value-weighted index of NYSE-listed stocks
MP = monthly growth rate in U.S. industrial production
DEI = change in inflation; measured by the U.S. consumer price index
UI = difference between actual and expected levels of inflation
UPR = unanticipated change in the bond credit spread (Baa yield − RFR)
UTS = unanticipated term structure shift (long-term less short-term RFR)

28
 Multifactor Models in Practice
 Burmeister, Roll, and Ross (1994)
analyzed the predictive ability of
a model based on the following
set of macroeconomic factors:
• Confidence risk
• Time horizon risk
• Inflation risk
• Business cycle risk
• Market timing risk

29
 Multifactor Models in Practice

 Fama and French (1993) developed a multifactor

model specifying the risk factors in microeconomic
terms using the characteristics of the underlying
securities
𝑅𝑖𝑡 − 𝑅𝐹𝑅𝑡 = 𝛼𝑖 + 𝑏𝑖1 𝑅𝑀𝑡 − 𝑅𝐹𝑅𝑡 + 𝑏𝑖2 𝑆𝑀𝐵𝑡 +𝑏𝑖3 𝐻𝑀𝐿𝑡 + 𝑒𝑖𝑡

• SMB (small minus big) = return to a portfolio of small capitalization

stocks less the return to a portfolio of large capitalization stocks
• HML (high minus low) = return to a portfolio of stocks with high ratios
of book-to-market values less the return to a portfolio of low book-to-
market value stocks

30
 Multifactor Models in Practice

 Carhart (1997), based on the Fama-French three-

factor model, developed a four-factor model by
including a risk factor that accounts for the tendency
for firms with positive past return to produce positive
future return

( Rit − RFRt ) =  i + bi1 ( RMt − RFRt ) + bi 2 SMBt + bi 3 HMLt + bi 4 MOM t + eit

Where
MOMt = the momentum factor

31
 Multifactor Models in Practice
 Fama and French (2015) developed their own extension
of the original three-factor model by adding two
additional terms to account for company quality: a
corporate profitability risk exposure and a corporate
investment risk exposure
( Rit − RFRt ) =  i + bi1 ( RMt − RFRt ) + bi 2 SMBt + bi3 HMLt + bi 4 RMWt + bi 5CMAt + eit
Where
RMW (robust minus weak) = return to a portfolio of high profitability
stocks less the return to a portfolio of low profitability stocks
CMA (conservative minus aggressive) = return to a portfolio of stocks
of low-investment firms (low total asset growth) less the return to a
portfolio of stocks in companies with rapid growth in total assets

32
 Multifactor Models in Practice
 Extensions of Characteristic-Based Risk Factor Models
• One type of security characteristic-based method for
defining systematic risk exposures involves the use of
index portfolios (e.g. S&P 500, Wilshire 5000) as common
risk factors, such as the one by Elton, Gruber, and Blake
(1996), who rely on four indexes:
• The S&P 500
• The Barclays Capital aggregate bond index
• The Prudential Bache index of the difference between large- and
small-cap stocks
• The Prudential Bache index of the difference between value and
growth stocks

33
Multifactor Models in Practice

34
Multifactor Models in Practice

35
 Single Index Model
From Investments by Bodie, Kane, and Marcus
 Single-Index Model

 Regression equation:
Ri (t ) =  i + i RM (t ) + ei (t )

 Expected return-beta relationship:

E (Ri ) =  i +  i E (RM )

37
 Single-Index Model

 Variance = Systematic risk + Firm-specific risk:

 i2 =  i2 M2 +  2 (ei )

 Covariance = Product of betas × Market risk:

Cov ( ri , rj ) = i  j M2

38
 Single-Index Model

 Correlation
i  j M2  i M2  j M2
Corr ( ri , rj ) = =
 i j  i M  j M
= Corr ( ri , rM )  Corr ( rj , rM )

39
Single-Index Model

40
Single-Index Model

41
Index Model Regression Equation
Excess return of security i

Ri ( t ) =  i + i RS & P 500 ( t ) + ei ( t )
Zero-mean, firm-
Expected excess
specific surprise in
return when the
security i‘s return in
market excess
month t.
return is zero
(the residual)
Sensitivity of
security i‘s return Expected excess
to changes in the return of the
return of the market
market

42
Index Model Regression Equation

43
 Interpreting the Output

 Correlation of Ford with the S&P 500 is 0.6280

 The model explains about 38% of the variation in
Ford
 Ford's alpha is -0.98% per month, but not statistically
significant
 Ford's beta is 1.3258, but the 95% confidence interval
is 0.90 to 1.75

44
 Portfolio Construction and
the Single-Index Model
 Alpha and Security Analysis
• Macroeconomic analysis estimates the risk premium and
market risk
• Statistical analysis estimates the beta coefficients and
residual variances, σ2(ei), of all securities
• Establish the expected return of each security absent any
contribution from security analysis
• Use security analysis to develop private forecasts of the
expected returns for each security

45
 Portfolio Construction and
the Single-Index Model
 Single-Index Model Input List
• Risk premium on the S&P 500 portfolio
• Estimate of the SD of the S&P 500 portfolio
• n sets of estimates of
• Beta coefficient
• Stock residual variances
• Alpha values

46
 Portfolio Construction and
the Single-Index Model
 Optimal risky portfolio in the single-index model
• Expected return, SD, and Sharpe ratio:

47
Portfolio Construction: The Process

 Optimal risky portfolio in the single-index model is a

combination of
• Active portfolio, denoted by A
• Passive portfolio, denoted by M

48
Summary of Optimization Procedure

 Compute the initial position of each security:

i
w =
0
i
 2 (ei )

 Scale those positions:

wi0
wi = n

 i
w 0

i =1

 Compute the alpha of the active portfolio:

n
 A =  wi i
i =1

49
Summary of Optimization Procedure

 Compute the residual variance of A:

n
 2 (eA ) =  wi2 2 (ei )
i =1

 Compute the initial position in A:

  2
(e A )
wA =
0 A

E ( RM )  M2
 Compute the beta of A:
n
 A =  wi i
i =1

50
Summary of Optimization Procedure

 Adjust the initial position in A:

0
w
w*A = A
1 + (1 −  A )  wA0

Note when  A = 1
→ w*A = wA0

51
Summary of Optimization Procedure

 Optimal Risky Portfolio now has weights:

wM = 1 − wA
wi = wA  wi
 Calculate the risk premium of P (Optimal Risky
Portfolio):
E ( RP ) = ( wM + wA  A )  E ( RM ) + wA A

 Compute the variance of P:

 2 (eA ) = ( wM + wA  A ) 2  M2 + [ wA (eA )]2

52
 Optimal Risky Portfolio:
Information Ratio
 The Information Ratio
• The contribution of the active portfolio depends on the ratio
of its alpha to its residual standard deviation:
Information Ratio of A
  2
(e A )
wA =
0 A

E ( RM )  M2
 The information ratio measures the extra return we
can obtain from security analysis

53
 Optimal Risky Portfolio:
Sharpe Ratio
 The Sharpe ratio of an optimally constructed risky
portfolio will exceed that of the index portfolio:
2
 A 
s P = s M +  (eA ) 
2 2

54
Summary of
Optimization
Process

55