Chapter 6

Introduction to Return and

Risk
Road Map

Part A Introduction to Finance.

Part B Valuation of assets, given discount rates.

Part C Determination of risk-adjusted discount rates.

Introduction to return and risk.

Portfolio theory.
CAPM and APT.

Part D Introduction to derivative securities.

Main Issues
Dening Risk

Estimating Return and Risk

Risk and Return - A Historical Perspective

Chapter 6 Introduction to Return and Risk 6-1

1 Asset Returns
Asset returns over a given period are often uncertain:

1 + P1 P0
D 1 + P1
D
r = = 1
P0 P0

where

denotes an uncertain outcome (random variable)

P0 is the price at the beginning of period

P1 is the price at the end of period - uncertain

1 is the dividend at the end of period - uncertain.
D

Return on an asset is a random variable, characterized by

all possible outcomes, and

probability of each outcome (state).

Example. The S&P 500 index and the stock of MassAir, a

regional airline company, give the following returns:

State 1 2 3
Probability 0.20 0.60 0.20
Return on S&P 500 (%) -5 10 20
Return on MassAir (%) -10 10 40

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-2 Introduction to Return and Risk Chapter 6

Risk in asset returns can be substantial.

Monthly Returns - IBM (1990 2000)

Monthly Returns of IBM from 1990 to 2000
0.3

0.2

0.1

0
Return

0.1

0.2

0.3

0.4
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Month

Annual Returns - S&P 500 Index (1926 2004)

Return on S&P

60.00%

40.00%

20.00%

0.00%

-20.00%

-40.00%

-60.00%
26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 01
19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-3

Expected rate of return on an investment is the discount rate

for its cash ows:
E0[D 1 + P1 ]
r E[r] = 1
P0
or
E0[D 1 + P1 ]
P0 =
1 + r
where denotes an expected value.

Expected rate of return compensates for time-value and risk:

r = rF +
where rF is the risk-free rate and is the risk premium
- rF compensates for time-value
- compensates for risk.

Questions:

1. How do we dene and measure risk?

2. How are risks of dierent assets related to each other?

3. How is risk priced (how is determined)?

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-4 Introduction to Return and Risk Chapter 6

2 Defining Risk
Example. Moments of return distribution. Consider three assets:

Mean StD
r0 (%) 10.0 0.00
r1 (%) 10.0 10.00
r2 (%) 10.0 20.00

Probability Distribution of Returns

4

3.5
riskless return of 10%

risky return of mean 10%

2.5 and volatility 10%
probability

risky return of mean 10%

and volatility 20%
1.5

0.5

0
0.4 0.2 0 0.2 0.4 0.6
return

Between Asset 0 and 1, which one would you choose?

Between Asset 1 and 2, which one would you choose?

Investors care about expected return and risk.

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-5

Key Assumptions On Investor Preferences for 15.401

1. Higher mean in return is preferred:

r = E[
r ].

2. Higher standard deviation (StD) in return is disliked:


= E[( r )2].
r

3. Investors care only about mean and StD (or variance).

Under 1-3, standard deviation (StD) gives a measure of risk.

Investor Preference for Return and Risk

Expected return (r)

6
6
@
I
@
@
@
@
@
@
@
@
@ increasing return
@
@
@
 @
decreasing risk
-
Risk ( )

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-6 Introduction to Return and Risk Chapter 6

3 Historical Return and Risk

Three central facts from history of U.S. nancial markets:

1. Return on more risky assets has been higher on average than

return on less risky assets:

Average Annual Total Returns from 1926 to 2005 (Nominal)

Asset Mean (%) StD (%)

T-bills 3.8 3.1
Long term T-bonds 5.8 9.2
Long term corp. bonds 6.2 8.5
Large stocks 12.3 20.2
Small stocks 17.4 32.9
Ination 3.1 4.3

Average Annual Total Returns from 1926 to 2005 (Real)

Asset Mean (%) StD (%)

T-bills 0.7 4.0
Long term T-bonds 2.9 10.4
Long term corp. bonds 3.2 9.7
Large stocks 9.1 20.3
Small stocks 13.9 32.3

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-7

Return Indices of Investments in the U.S. Capital Markets

Real returns from 1926 to 2004

Security Initial Total Return
T-Bills $1.00 1.74
Long Term T-Bonds $1.00 6.03
Corporate Bonds $1.00 8.86
Large Stocks $1.00 242.88
Small Stocks $1.00 1,208.84

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-8 Introduction to Return and Risk Chapter 6

2. Returns on risky assets can be highly correlated to each other:

Cross Correlations of Annual Nominal Returns (1926 2005)

Bills T-bonds C-bonds L. stocks S. stocks Ination

T-bills 1.00 0.23 0.20 -0.02 -0.10 0.41
L.T. T-bonds 1.00 0.93 0.12 -0.02 -0.14
L.t. C-bonds 1.00 0.19 0.08 -0.15
Large stocks 1.00 0.79 -0.02
Small stocks 1.00 0.04
Ination 1.00

Cross Correlations of Annual Real Returns (1926 2005)

Bills T-bonds C-bonds L. stocks S. stocks

T-bills 1.00 0.57 0.57 0.11 -0.06
L.T. T-bonds 1.00 0.95 0.20 0.02
L.t. C-bonds 1.00 0.26 0.11
Large stocks 1.00 0.79
Small stocks 1.00

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-9

3. Returns on risky assets are serially uncorrelated.

Serial Correlations of Annual Asset Returns (1926 2005)

Serial Correlation
Asset Nominal return Real return
T-bills (risk-free) 0.91 0.67
Long term T-bonds -0.08 0.02
Long term corp. bonds 0.08 0.19
Large stocks 0.03 0.02
Small stocks 0.06 0.03

(Note: The main source for the data in this subsection is Stocks, bonds, bills and inflation,

2006 Year Book, Ibbotson Associates, Chicago, 2006.)

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-10 Introduction to Return and Risk Chapter 6

4 Risk and Horizon

Previous discussions focused on return and risk over a xed
horizon. Often, we need to know:

How do risk and return vary with horizon?

How do risk and return change over time?

We need to know how successive asset returns are related.

IID Assumption: Asset returns are IID when successive returns

are independently and identically distributed.

For IID returns, r1, r2, . . . , rt are independent draws from the
same distribution.

Pt is the asset price (including dividend). The continuously

compounded return is

Pt Pt
= ert or log = log Pt log Pt1 = rt.
Pt1 Pt1

Denition: Asset price (in log) follows a Random Walk when its
changes are IID.

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-11

Example. An IID return series a binomial tree for prices:

115.76
110.25
107.49
105
107.49
102.38
99.82
P0 =100
107.49
102.38
99.82
97.5
99.82
95.06
92.68

where

(1) price can go up by 5% or down by 2.5% at each node

(2) probabilities of up and down are the same at each node.

For the above binomial price process:

Successive returns are independent and identically distributed.

If up and down are equally likely, expected return is

(log 1.05 + log 0.975)/2 = 1.17%.

Return variance for one-period is

 2
2= 1 1.05
1 log = (0.0371)2 .
2 0.975

Return variance over T periods is (0.0371)2 T .

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-12 Introduction to Return and Risk Chapter 6

Implications of the IID assumption

(a) Returns are serially uncorrelated.

(b) No predicable trends, cycles or patterns in returns.

(c) Risk (measured by variance) accumulates linearly over time:

Annual variance is 12 times the monthly variance.

Advantage of IID assumption:

Future return distribution can be estimated from past returns.

Return distribution over a given horizon provides sucient

information on returns for all horizons.

IID assumption is consistent with information-ecient market.

Weakness of IID assumption:

Return distributions may change over time.

Returns may be serially correlated.

Risk may not accumulate linearly over time.

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-13

5 Investment in the long-run

Are stocks less risky in the long-run? Not if returns are IID.
Higher expected total return.
Higher probability to outperform bond.
More uncertainty about terminal value.

Example. Return proles for dierent horizons.

rbond = 5%.
rstock = 12% and stock = 20%.

Probability Distribution of Terminal Value in Two years Probability Distribution of Terminal Value in Five Years

1.2 Probability of stock beating bond = 0.6236 1.2 Probability of stock beating bond = 0.6907

1 1
bond return (gross) = 1.1025

bond return (gross) = 1.2763

0.8 0.8
Probability

Probability

0.6 0.6

0.4 mean stock return (gross) = 1.2544 0.4

mean stock return (gross) = 1.7623

0.2 0.2

0 0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
Terminal value of one dollar investment in stock and bond Terminal value of one dollar investment in stock and bond

Probability Distribution of Terminal Value in Ten Years

1.2 Probability of stock beating bond = 0.7594

0.8
bond return (gross) = 1.6289
Probability

0.6

0.4

mean stock return (gross) = 3.1058

0.2

0
0 1 2 3 4 5 6 7 8
Terminal value of one dollar investment in stock and bond

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-14 Introduction to Return and Risk Chapter 6

6 Appendix: Probability and Statistics

Consider two random variables: x
and y

State 1 2 3 n
Probability p1 p2 p3 pn
Value of x x1 x2 x3 x n
Value of y y1 y2 y3 yn
n
where i=1 pi = 1.

1. Mean: The expected or forecasted value of a random outcome.

n

E[
x] = x
= pj xj .
j=1

2. Variance: A measure of how much the realized outcome is likely to dier

from the expected outcome.
n

 
2
x] = x2 = E (
Var[ xx
) 2 = pj xj x
.
j=1
Standard Deviation (Volatility):

StD[
x] = x = Var[
x].

3. Skewness: A measure of asymmetry in distribtion.

  
3
Skew[
x] = )3 /x .
E (x x

4. Kurtosis: A measure of fatness in tail distribution.

  
4
Kurtosis[
x] = E )4 /x .
(x x

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-15

Example 1. Suppose that random variables x

and y are the returns on S&P
500 index and MassAir, respectively, and

State 1 2 3
Probability 0.20 0.60 0.20
Return on S&P 500 (%) -5 10 20
Return on MassAir (%) -10 10 40

Expected Value:

x
= (0.2)(0.05) + (0.6)(0.10) + (0.2)(0.20) = 0.09

y = 0.12

Variance:

x2 = (0.2)(0.050.09)2 +
(0.6)(0.100.09)2 +
(0.2)(0.200.09)2
= 0.0064

y2 = 0.0256

Standard Deviation (StD):

x = 0.0064 = 8.00%

y = 16.00%.

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-16 Introduction to Return and Risk Chapter 6

Covariance and Correlation

1. Covariance: A measure of how much two random outcomes vary to-

gether.
Cov[ xx
x, y] = xy = E [( ) (
y y)]
n


= pj xj x
yj y .
j=1

2. Correlation: A standardized measure of covariation.

xy
Corr[x, y] = xy = .
x y

Note:
(a) xy must lie between -1 and 1.

(b) The two random outcomes are

Perfectly positively correlated if xy = +1
Perfectly negatively correlated if xy = 1
Uncorrelated if xy = 0.

(c) If one outcome is certain, then xy = 0.

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-17

Example 1. (Continued.) For the returns on S&P and MassAir:

State 1 2 3
Probability 0.20 0.60 0.20
Return on S&P 500 (x) (%) -5 10 20
y ) (%)
Return on MassAir ( -10 10 40

with mean and StD:

x
= 0.09, x = 0.08,

y = 0.12, y = 0.16.

We have

Covariance:

xy = (0.2)(0.050.09)(0.100.12) +
(0.6)(0.100.09)(0.100.12) +
(0.2)(0.200.09)(0.400.12)
= 0.0122.

Correlation:
0.0122
xy = = 0.953125.
(0.08)(0.16)

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-18 Introduction to Return and Risk Chapter 6

Computation Rules

Let a and b be two constants.

E[a
x] = a E[
x].

E[a
x +b
y ] = a E[
x] + b E[
y ].

E[
xy] = E[
x] E[
y ] + Cov[
x, y].

x] = a2 Var[
Var[a x] = a2 x2 .

Var[a y ] = a2 x2 + b2 y2 + 2(ab)xy .
x +b

Cov[
x + y, z] = Cov[
x, z] + Cov[
y , z].

Cov[a
x, b
y ] = (ab)Cov[
x, y] = (ab)xy .

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-19

Linear Regression

Relation between two random variables y and x

:

y = +
x+


where

Cov[
y, x
] yx
= = 2
Var[
x] x
= y
x

Cov[
x,
] = 0.

gives the expected deviation of y from y for a given deviation of x

from
x
.

 has zero mean: E[
] = 0 .

 represents the part of y that is uncorrelated with x:
Cov[
x,
] = 0.

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-20 Introduction to Return and Risk Chapter 6

Furthermore:

y2 = Var[
y ] = Var[+
x +
]

= 2 x2 + 2

Total Variance = Explained Variance

+ Unexplained Variance.

Explained variance: 2 x2

Unexplained variance: 2 .

What fraction of the total variance of y is explained by x

?

Explained Variance 2 x2 2 x2
R2 = = = .
Total Variance y2 2 x2 + 2

15.401 Lecture Notes 

c J. Wang Fall 2006
Chapter 6 Introduction to Return and Risk 6-21

Example 1. (Continued.) In the above example: x

is the return on S&P 500
and y is the return on MassAir.

0.0122
= = 1.9062 and = 0.0516.
0.082

State 1 2 3
Probability 0.20 0.60 0.20
Return on S&P 500 (%) - 5.00 10.00 20.00
Return on MassAir (%) -10.00 10.00 40.00

 = y ( +
x) (%) 4.69 - 3.90 7.03

Moreover,

x2 = 0.0064, y2 = 0.0256, 2 = 0.00234

and

(1.9062)2 (.0064)
R2 = = 0.9084
(.0256)

1 R2 = 0.0916.

Fall 2006 
c J. Wang 15.401 Lecture Notes
6-22 Introduction to Return and Risk Chapter 6

7 Homework
Readings:

BKM Chapter 5.25.4.

BMA Chapter 7.1.

15.401 Lecture Notes 

c J. Wang Fall 2006