ARBITRAGE PRICING

THEORY & SHARPE

INDEX MODEL
Rahul Savalia
Mithilesh Shukla
Arbitrage Pricing Theory (APT)
Based on the law of one price. Two items that
are the same cannot sell at different prices
If they sell at a different price, arbitrage will
take place in which arbitrageurs buy the good
which is cheap and sell the one which is
higher priced till all prices for the goods are
equal
APT
In APT, the assumption of investors utilizing a
mean-variance framework is replaced by an
assumption of the process of generating
security returns.
APT requires that the returns on any stock be
linearly related to a set of indices.
In APT, multiple factors have an impact on the
returns of an asset in contrast with CAPM
model that suggests that return is related to
only one factor, i.e., systematic risk
Factors that have an impact the returns of all
assets may include inflation, growth in GNP,
major political upheavals, or changes in
interest rates
r
i
= a
i
+ b
i1
F
1
+ b
i2
F
2
+ +b
ik
F
k
+ e
i
Given these common factors, the b
ik
terms
determine how each asset reacts to this
common factor.
While all assets may be affected by growth in
GNP, the impact will differ.
Which firms will be affected more by the
growth in GNP?
The APT assumes that, in equilibrium, the
return on a zero-investment, zero-systematic
risk portfolio is zero, when the unique effects
are diversified away:
E(r
i
) =
0
+
1
b
i1
+
2
b
i2
+ +
k
b
ik



Single Index Model-
Assumptions
Known economist William Sharpe developed
the single index model.
According to this model we make following
assumptions:
a) We summarize all relevant economic factors
by one macro-economic indicator and assume
that it moves the security market as a whole.
b) Beyond this common effect, all remaining
uncertainty in stock return is firm specific; i.e.
there is no other source of correlation
between securities.
c) Stocks co vary together only because of their
common relationship to the market index.
Single Index Model
This model relates returns on each security to
the returns on a common index.
A broad index of common stock is generally
used for this purpose.
The common index can be BSE 100 stocks,
Nifty 50 stocks so on and so forth.

Single Index Model- Returns
The single index model can be expressed by the following
equation.



R
i
= the return on security i
R
M
=the return on the market index

i
=that part of security is return independent of market
performance.

i
= a constant measuring the expected change in the
dependent variable, Ri, given a change in the independent
variable RM
e
i
= random residual error.

i M i i i
e R R + + = | o
Single Index Model- Returns
Single index model divides return into two
components
1. a unique part,
i
2. a market-related part,
i
R
M
The unique part is a micro event, affecting an
individual company but not all companies in
general.
The market related part, on the other hand, is
a macro event that is broad based and
affects all (or most) of the firms.
The error term e
i
captures the difference
between the return that actually occurs and
return expected to occur given a market
index return.
Measuring Systematic Risk
Beta () measures a stocks market (or
systematic) risk. It shows the relative volatility of
a given stock compared to the average stock.
An average stock (or the market portfolio) has a
beta = 1.0.
Beta shows how risky a stock is if the stock is
held in a well-diversified portfolio.

=1 stock has average risk.
>1 stock is riskier than average.
<1 stock is less risky than average.
=0 risk free assets (e.g., Treasury bills)


Single Index Model- Co
variance
In the single index model, the covariance between two
stocks depends only on the market risk Therefore
covariance between two securities can be written as


Note that stock is beta has two components:
Covariance of returns between stock i and market
portfolio.
Variance of return on market portfolio
NO variance of return on stock i


2

M i ij
o | o =
NO o
i
2

Single Index Model- Risk
In Single Index Model, the total risk of a
security, as measured by its variance,
consists of two components: market risk and
unique risk



= Market risk+ company specific risk

2 2 2 2

ei M i
i
o o | o + =
Single Index Model- Risk
This single security variance can be
extrapolated for finding the minimum variance
set of portfolios.


i.e. Total portfolio variance=Portfolio market
risk+ portfolio residual variance
2 2 2 2
ep M p p
o o | o + =
Reward to Risk Ratio
We can vary the amount invested in each type
of asset and get an idea of the relation
between portfolio expected return and beta:



It estimates the expected risk premium per unit
of risk.
We can also calculate the reward to risk ratio
for all individual securities.
P
f P
R R E
|

=
) (
Ratio Risk - to - Reward
What happens if two securities have
different reward-to-risk ratios?


Investors would only buy the securities (portfolios) with a higher
reward-to-risk ratio. Here, it would be A.
Eventually, all securities will have the same reward-to-risk ratio.
Because the reward-to-risk ratio is the same for all securities, it
must hold for the market portfolio too.
Result:

B
f B
A
f A
R R E R R E
| |

>
) ( ) (
f M
M
f M
B
f B
A
f A
R R E
R R E R R E R R E
=

= =

) (
) (
...
) ( ) (
| | |
Thank You