Chapter 3: Asset

Valuation Models
Capital Asset Valuation Model (CAPM)
• Multi-Factor Model and Arbitrage Pricing
Theory (APT)

2
 Capital Asset Pricing Model -
CAPM
This is the central valuation model of modern financial economic
theory.
The model gives us an accurate forecast of the relationship that
needs to be observed between an asset's risk and its expected
return.
• This relationship is manifested in two essential functions. Firstly, it
provides a standard rate of return for valuing investment
opportunities. Second, the model helps us make a return-
oriented estimate of expectations on assets that have not yet
been traded on the secondary market.

3
 CAPM
This is the equilibrium model that underlies all
modern financial theory.
This model is built from the use of distributed principles with
simplified assumptions.
Markowitz, Sharpe, Lintner and Mossin are credited with
developing this model.
• CAPM is a balanced pricing model. In particular,
equilibrium is understood: A situation when the price of a
stock (or the expected rate of return) represents the
"consensus" of the market. No investor wants to change
their holding (buying, or selling) securities at a balanced
price.

4
 CAPM (tt)
Hypothesis
There are many investors, each individual investor cannot
influence the price of securities through his or her trading
behaviors. In other words, the investor only follows the price
of the market and is not capable of influencing the price. This
is the "perfect market" hypothesis in microeconomics.
All investors have the same investment period.
Investment activities are limited to traded financial assets.
• There are no taxes and transaction fees. Investors can
borrow and lend in unlimited quantities with risk-free
interest rates.

5
 CAPM (tt)
Hypothesis
All investors are rational and wise investors by diversifying their
investment according to the Markowitz model (on the basis of
expected return – standard deviation).
Information is inexpensive and available to all investors.
• Investors will have the same estimate of the expected return on
the stocks, the standard deviation of the yield rate, and the
covariance matrix. Therefore, investors have the same estimate
of the effective border. This assumption is also known as "uniform
expectation".

6
Consequences of equilibrium conditions
Conclusion 1
E(R)

CAL => CML

P =>M
E(RM)
RF

M
7
Consequences of equilibrium
conditions Conclusion
1 (tt)
• Conclusion 1: All investors hold the same optimal risk
portfolio: market portfolio (M).
– Market Categories(M):
– Contains all types of stocks traded in the market.
– The proportion of each type of stock = the market value of each
type
– CK/total market value of all types of stocks.
– TT value of CK = CK price x Number of CK in the market
• Thus, if stock A accounts for 1% of the market, the proportion of
A in the investment fund of each investor is also 1%.
8
 Conclusion
1(tt)
• The M market portfolio includes all risk assets in the economy?
– Suppose if investors' optimal portfolio of risk assets does not include
B shares, then the demand for B shares is 0. The price of B stock will
fall continuously until B stock becomes more attractive than other
stocks. Finally, B shares will also be included in investors' portfolios.
• The market portfolio is not only on the effective border, but also
tangent to the optimal capital allocation line (CAL).
• Capital Market Lines (CML):
– is a straight line connecting 𝑅𝐹 and M
– is the optimal capital allocation line (CAL) (coefficient Sharp  Max)
– demonstrates a balanced relationship between the expected yield
and the risk of effective portfolios.

9
 Capital Market Line (CML)

• CML line equation:

𝐸 𝑅𝑀 −
𝐸 𝑅𝐶 = 𝑅𝐹
𝑅𝐹
+ 𝜎𝐶
In which: 𝜎𝑀
• [𝐸 𝑅𝑀 − 𝑅𝐹] ∶ Market risk
compensation
• 𝐸 𝑅𝑀
𝜎
−𝑅𝐹
: CML road angle coefficient

• 𝑅𝐹 𝑀 : CML Road Blocking

∶ Coefficient

10
 Capital Market Road (tt)
E(R)

CML
M
E(RM)
RF

M
11
Capital Market Lines
with the assumption of
borrowing or lending at a risk-
free interest rate

12
• The only question remains: At what price
will investors be willing to hold a risky asset
in their optimal risk portfolio?

13
 Conclusion 2
• The risk compensation for individual asset holdings
will be proportional to the risk compensation of the
DMTT (M) and the β ratio of the security.
𝐸 𝑅𝐴 − 𝑅𝐹 =𝐸 𝑅𝑀 − 𝑅𝐹 𝛽𝐴
• The β coefficient measures the variable
relationship in returns between the fluctuation of
the value of a stock relative to the fluctuation of

�= 𝐶𝑜𝑣
the market.
�
𝜎�
� 𝑅 𝐴 ,𝑅 𝑀
2�
�

14
 Conclusion 2 (continue)
• The risk compensation of the market portfolio (M) is:

𝐸 𝑅𝑀 − 𝑅𝐹 = ∑ 𝑤𝑖𝐸 𝑅𝑖 −∑
𝑤𝑖𝑅𝐹 = ∑ 𝑤𝑖[𝐸 𝑅𝑖 − 𝑅𝐹]

market portfolio (M):𝑤𝐴 𝐸 𝑅𝐴 − 𝑅𝐹

 Contribution of asset (A) to the risk compensation of the

 The contribution of a security to the risk compensation of

the M market portfolio is equal to the proportion of that
asset in the market portfolio multiplied by the risk
compensation of the asset itself.
15
 Properties of Covariance
1. Cov(X,X) = var(X)
2. Cov(X,Y) = Cov(Y,X)
3. Cov(aX,Y) = a Cov(X,Y) = Cov(X,aY)
4. Cov(X,Y+Z) = Cov(X,Y) + Cov(X,Z)
with X,Y,Z = variables and a
= const.

16
• Cov(𝑅A, 𝑅𝑀)
= Cov(𝑅A, w1𝑅1 + w2𝑅2 + ⋯ + w𝐴𝑅𝐴+…
+w𝑛𝑅𝑛)
Cov(𝑅A, w1𝑅1) + Cov(𝑅A,
w2𝑅2) + …+ Cov(𝑅A, w𝐴𝑅𝐴) + …+
=

Cov(𝑅A, w𝑛 𝑅 𝑛 )
= w1Cov(𝑅A, 𝑅1) + w2Cov(𝑅A, 𝑅2)
+ …+
w𝐴Cov(𝑅A, 𝑅𝐴) + …+ w𝑛Cov(𝑅A, 𝑅𝑛)
17
 Matrix of Variance - Covariance

Tỷ trọng w1 w2 … wA … wn

w1 cov(R1,R1) cov(R1,R2) … cov(R1,RA) … cov(R1,Rn)

w2 cov(R2,R1) cov(R2,R2) … cov(R2,RA) … cov(R2,Rn)

… … … … … … …

… … … … … … …

wA cov(RA,R1) cov(RA,R2) … cov(RA,RA) … cov(RA,Rn)

… … … … … … …

wn cov(Rn,R1) cov(Rn,R2) … cov(Rn,RA) … cov(Rn,Rn)

18
• The variance (risk) of the market portfolio (M) is
determined:

In which:
The contribution of asset (A) to the risk of (M) is:
wA× Cov(RA,RM)
• The contribution of a security to the risk of the M market
portfolio is equal to the proportion of that asset in the market
portfolio multiplied by the covariance of the return of that
security with the other assets that make up the market
portfolio.
19
 Reward/Risk Ratio
=
𝑤 𝐴 𝐸(𝑅 𝐴 −𝑅 𝐹 ] 𝐸(𝑅 𝐴
𝑤 𝐴 Cov(𝑅 𝐴 ,𝑅 Cov(𝑅
−𝑅𝐹] 𝐴 ,𝑅
• For Assets(A):
𝑀 )𝐸(𝑅 𝑀 −𝑅 𝐹 ] 𝑀)
• For DMTT(M):
Cov(𝑅 𝑀 ,𝑅 𝐸(𝑅
σ𝑀2
=
𝑀)
�

• The market is balanced, therefore:−𝑅𝐹]

�

𝐸(𝑅𝐴 − 𝑅𝐹]
=�
σ
𝐸(𝑅𝑀 − 𝑅𝐹] 2 �
 𝐸(𝑅𝐴) =
Cov 𝐸( � − 𝑅�
− 𝑅𝐹 Cov(𝑅
𝑅 𝐴 ,𝑅𝐴
,
σ 𝑅𝑀) 𝑅 � ] �
2

 𝐸(𝑅𝐴) − 𝑅𝐹 𝐸(𝑅 −
�
𝑀 �

= 𝛽𝐴 𝑀 𝑅𝐹] 20
 Example
• E(RM)-RF = 0.08; RF = 0.03

• x = 1.25
E(RX) = 0.03 + 1.25(0.08) = 0.13 or 13%

• Y = 0.6
E(RY) = 0.03 + 0.6(0.08) = 0.078 or 7.8%

21
 Beta Factor
 𝐴,𝑀
𝐴 = 𝐶𝑜𝑣 σ𝐴
= �
�
𝑀
𝜎 2𝑅 𝐴 ,𝑅 𝑀 𝑀
� �
The systemic risk of a security is expressed in the
beta coefficient. This coefficient measures the
volatility of a security compared to the volatility of the
stock market index.
• The higher the beta ratio of a security, the greater
the systemic risk of this security.

22
 Beta Factor (continue)

• Risk-free securities (F), e.g. Treasury bills, have a

coefficient βF = 0.
• Market category (M) with coefficient βM = 1.
• If an asset has β > 1, it has more systemic risk than
the market portfolio (M).
• If an asset has β < 1, it has less systemic risk than the
market portfolio (M).
• The majority of stocks have a beta ratio that varies
from 0.5 to 2.
23
 Portfolio Beta Factor
• Portfolio (P) consists of n Assets
n
•
P  w11  w22  ...  wnn   wi i

– 𝑤𝑖 : proportion of Assets (i) in the portfolio

i 1
In which:

(P)
– β𝑖 : Asset Beta Factor (i)

24
Portfolio Beta Ratio
(continue)

25
 SML
Because the SML simulates the relationship
between expected returns - beta coefficients, the
right valuation securities must be above the SML
line.
Undervalued securities will provide returns
that exceed expected returns, and therefore, are
above the SML line.
• Conversely, the overvalued securities will be
below the SML line.

26
 SML and CML
The CML specifies the risk compensation of the combined
investment funds (the portfolio includes non-risky assets and
the market portfolio), which is a function of the standard
deviation of the investment portfolio.
SML points out that the risk compensation of individual assets is
a function of asset risk. The measurement of risk is not based
on the standard deviation or variance of the asset; Instead, it
bases the asset's contribution to the market portfolio variance
(represented by the beta factor).
• The beta coefficient of a security measures its contribution to
the variance of the market portfolio. So, the expected risk
offset is a function of beta.

27
 Alpha coefficient
• The difference between the actual expected rate of
return (based on their current price, futures, and future
dividends) and the nominal expected rate of return (the
required ratmn e of return determined by CAPM) is
known as the alpha coefficient (α) of the security.
E(R)
SML
B
α
M
Directory
Market
RF

β 28
 Applications of CAPM
Provides a standard method of calculating the rate
of return for use in the valuation of securities.
Evaluate the performance of an investment fund.
• Evaluate a company's investment opportunities.
In other words, financial managers can use the
CAPM model to find out the internal rate of
return (IRR) required for the project.

29
 Shortcomings of CAPM
Tests of the CAPM model show that the beta ratios of individual
securities are unstable, while the beta ratios of portfolios are generally
stable and commensurate with trading volume.
Some recent evidence suggests the need to add variables that represent
other risks.
• In addition, some studies criticize the criteria and usefulness of this
model in evaluating portfolios because of its reliance on the market's
portfolio of risk assets that are often unavailable.

30
 Factor Model
• The realized rate of return on securities i is
expressed according to the factor model as follows:

31
 Factor Model (tt)
Giả thiết:
• The risk of a security is separated into 2 parts: Risk systems
and risks are the company's own.
• The system risk section can be classified into several general
risk factors Fj. In which:
– The general risk factor (represented by Fj) is the unexpected
(unexpected) events of a certain economic variable that have an
effect (on a wide range) on the returns of securities. These
unexpected events can be unexpected changes in inflation,
GDP, interest rates, fuel prices, stock market indices…
– Because the risk factor is an unexpected event – determined
by the difference between the actual value of the economic
variable and the expected value of that variable – the
expected value of the risk factor is 0: [E(Fj) = 0]
32
 Factor Model (tt)
• Important note: The factor model is only an
explanatory model for the process of forming
returns on securities, not a valuation model.
•
• What does the factor model mean?
• Identify the elements of the model?
• Estimation of factor β?

33
 Factor Model (tt)
• Portfolio Factor Model:

• In which:

34
 Factor Model (tt)
• Factor category: A special category that is well
diversified, and is built to have a coefficient of
βk=1 for factor k and β=0 for other factors.
• Examples of building a portfolio of factors:
– Suppose there are 2 risk factors, and the 2-factor
model of 3 categories A, B, and C has been estimated
as follows:

35
 Examples of
building a portfolio
of factors
• If you call wi is the proportion of each
corresponding category A, B, and C in the list of
factors, then:
• The first category of factors will be determined from
the system of equations:

• The result will be:

36
 Example of building
a portfolio of factors
(tt)
• The second category of factors will be determined
from the system of equations:

• The result will be:

37
 APT
• APT's hypotheses:
- The return on securities is described by the factor model:
𝑅𝑖 = 𝐸(𝑅𝑖) + 𝛽𝑖,1𝐹1 + 𝛽𝑖,2𝐹2 + ⋯ + 𝛽𝑖,𝑘𝐹𝑘 + ε𝑖
– All investors have the same assessment of the future
prospects of securities (as assumed by CAPM).
– Investors are people who want to maximize useful
value (as CAPM assumes).
– The market is perfectly competitive, allowing no
arbitrage opportunities to exist.

38
 APT (tt)
• In particular, APT does not assume the M market
category as in CAPM.
• APT Model:
𝐸(𝑅𝑖) = 𝑅𝐹 + 𝛽𝑖,1λ1 + 𝛽𝑖,2λ2 + ⋯ + 𝛽𝑖,𝑘λ𝑘
In which:
λ𝑘 : Risk compensation of factor k.
 APT implies that when 2 categories have the same
β coefficients corresponding to each factor, the
expected return of the 2 categories must be equal.
39
 APT (tt)
• Stock valuation and process arbitrage:
– For example, suppose there are two categories
of factors. The expected yield on the first and
second factor categories, respectively, is E(RP)
= 10% and E(RQ) = 12%. Risk-free interest
rates Rf = 4%.
– A category A has a coefficient that measures
sensitivity to the first factor β1=0,5 and the
coefficient of measuring sensitivity to the
second factor β2=0,75.
40
 APT (tt)
• According to APT, the expected yield on Category
A will be:
𝐸(𝑅𝐴) = 4% + 0.5 10% − 4%+ 0.712% −
4%
=13%
• If the expected rate of return is estimated on
category A to be 12% (instead of 13%) →
Arbitrage Opportunity
• How to Exploit Arbitrage Opportunities?
41
 APT Accreditation
• Difficulties in APT accreditation:
• APT did not specify which factors affect stock
returns.
– Similar to CAPM, APT predicts expected
returns.
Key findings:
- Supporting evidence exists for APT.
- There can be 3 to 5 factors that affect the
return of a stock and are valued.
42
 Applications of APT
APT has the same application as CAPM:
Provides a standard rate of return calculation
for use in securities valuation.
Evaluate the performance of an investment
fund.
• Evaluating a company's investment
opportunities.

43
Estimation of factor risk compensation
and factor β
• Factor risk compensation:
- Usually estimated from the historical average risk
offset of the factor portfolio.
• Coefficient β factor:
- Regression of the return of securities to the return
of the portfolio of factors (using a period of historical
data). The angle coefficient of the regression model
is β factor.

44
 Some conclusions about asset
valuation models
• Is APT at odds with CAPM?
Which model is better?
- There is no specific answer yet, more experimental research
is needed.
- However, there is one consensus: determining which model is
better is not a very important requirement. Each model has a
solid foundation in relation to the balance in the market: For
APT, it is Arbitrage activity; For CAPM, it is optimal to hold an
effective portfolio.

45