200488 CORPORATE FINANCIAL MANAGEMENT

Workshop: WEEK 3

 EXPECTED RETURN AND RISK

 DIVERSIFICATION AND RISK

Questions to be considered:

 How do we measure expected return and risk?

 How do we calculate the expected return for a
portfolio?
 What is the distinction between systematic and
non-systematic risk?
 How does diversification affect the risk of a
portfolio?
 Why, in theory, is non-systematic risk irrelevant in
the valuation of risky assets?

1
EXPECTED RETURN AND RISK
Any financial decision involves some type of risk.
Recently we are experiencing a volatile stock market
due the uncertainty in the Chinese economy. The stock
market is volatile and the stock price plunged worldwide.
For example, a company borrows money faces interest
rate risk, which may change due to change in monetary
policy.

Therefore it is important to know risk and return to make

effective financial decision. This will help an investor to
maximize the expected return.

Risk is broadly defined as the chance of incurring a loss.

In the financial context it can be defined more
specifically as the probability of earning a return less
than the expected return.

2
(a) Measuring Expected Return and Risk for a
Single Asset

Consider the following return and risk data for ABC

Company.

Table 1.1
Rate of
ABC share R
Year Return i Probability, ( Pi )
price
(%)
2010 5.32
2011 6.54 22.93 0.2
2012 8.45 29.20 0.3
2013 8.56 1.30 0.3
2014 9.53 11.33 0.1
2015 8.66 -9.13 0.1
Total 1.00

Where,
 SPt ,i  SPt 1,i  Ct ,i 
Ri  Rate of return =    100%;
 SP t 1,i 
SPt ,i  Price of the investment 'i' (e.g., share, bond etc.) at time t.
SPt 1,i  Price of the investment 'i' (e.g., share, bond etc.) at time t-1.
Pi  Probability of each return (total probability must be 1.00)
Ct ,i  Dividend at time t for investment 'i'
(if divident is not given, consider it as zero)
Note: the probability of each return will follow a
probability distribution function (discrete or continuous-
for large number of observations)

3
Return is the financial outcome of an investment and
risk is the uncertainty associated with investment.
In general expected return on an investment can be
calculated as follows:
n
Expected return=E(R)= Ri Pi  R1P1  R2P2  R3P3  ....  RnPn
i 1
It is clear from the data that the investment return can
range from -9.13 to as high as 29.20. The spread of the
investment return is very high. This means that the
uncertainty of the return will also be high. There are
several measures of dispersion in statistics, such as:
Range, mean absolute deviation and the variance (also
coefficient of variation). Variance or standard deviation
(square root of variance) is the most accepted measure
of risk and calculated using the following formula:
n
Variance= =  Ri  E (R ) Pi
2 2

i 1

  R1  E (R ) P1   R2  E (R ) P2  ..   Rn  E (R ) Pn
2 2 2

Standard deviation= Variance  

4
The calculation of expected return and the risk, i.e.
variance is given in table 1.2

Table 1.2
Year Rate of
Probability,
Return
 Ri  E (R )
2
 Ri Pi  Pi
 Ri  (%)
( Pi )
2011 22.93 0.2 4.59 16.10
2012 29.20 0.3 8.76 69.72
2013 1.30 0.3 0.39 48.07
2014 11.33 0.1 1.13 0.69
2015 -9.13 0.1 -0.91 53.31
Sum 1.00 E(R)=13.96  2 =187.89

Therefore, in our example, the expected return for ABC

Company is 13.96% and corresponding standard
deviation is 187.89  13.7073

In general, the distribution of the returns follows discrete

distribution (such as Binomial, Poisson and Negative
Binomial etc.) depending on the nature of the data.
When n tends to infinity (or sufficiently large) the
distribution follows Normal distribution.

Note that the risk associated with the returns to an asset

is generally measured by the variability of returns to the
asset. If the return on an asset is constant (never
varies), we say the asset is risk-free. We see that the
variance is the ‘average squared distance’ between the
expected returns and the mean return.
An advantage of the standard deviation over the
variance is that it is expressed in the original units of
measure.

5
 Thus if the return to an asset is measured in percentage
terms, the standard deviation will also be measured as a
percent, whilst the variance will be expressed as
squared percent

Example:
Suppose we have the following discrete probability
distribution of returns on three different assets:
government bonds, ordinary shares of company A, and
ordinary shares of company B. Calculate the expected
returns and corresponding standard deviation for these
assets

% Returns for a period (e.g year)

Economic Probability Bonds Company A Company
Condition shares B shares
Very poor 0.1 9 -2 -3
Poor 0.2 8 3 2
Average 0.4 6 8 9
Good 0.3 4 11 13
1.0

For the above example:

For Bond:
EBond (R )  9 (0.1)  8 (0.2)  6 (0.4)  4 (0.3)  6.1%
For Company A:
E A (R )  2 (0.1)  3(0.2)  8 (0.4)  11(0.3)  6.9%
For Company B:
EB (R )  3 (0.1)  2 (0.2)  9 (0.4)  13 (0.3)  7.6%

6
  Bond   96.1 ( 0.1)  (8  6.1) 2 ( 0.2 )  ( 6  6.1) 2 ( 0.4 )  ( 4  6.1) 2 ( 0.3)
2

 2. 89  1.7%
 CompanyA   26. 9 ( 0.1)  ( 3  6. 9 ) 2 ( 0.2 )  (8  6. 9 ) 2 ( 0.4 )  (11  6. 9 ) 2 ( 0.3)
2

 16. 49  4.06%
 CompanyB   37.6 ( 0.1)  ( 2  7.6 ) 2 ( 0.2 )  ( 9  7.6 ) 2 ( 0.4 )  (13  7.6 )2 ( 0.3)
2

 27. 04  5.2%

We could also have worked in terms of decimal

proportions instead of percentages.

7
The Coefficient of Variation of Returns

CV   100%
The coefficient of variation E ( R ) is a
measure of relative variability used to compare the
levels of risks associated with returns for different
assets.

Example: Calculate the coefficients of variation for

the previous 3-asset example.
1.7
CV  100  27.87%
For bonds: 6.1
4.06
CV   100  0.58.84%
For company A: 6.9
5.2
CV  100  68.42%
For company B: 7.6

Thus the returns to Company B shares are relatively

more variable than the returns to bonds and Company A
shares.

The coefficient of variation shows risk per unit of

expected return.

8
As a more striking example suppose we have

Returns (%)
State of Nature Probability Company C Company D
shares shares
1 1/3 1 20
2 1/3 2 21
3 1/3 3 22

Mean 2 21
Std. Deviation 0.816 0.816
CV 0.408 0.039

(b) Measuring Expected Return and Risk when a

Probability Distribution of Returns is not known
or assumed

In this case we must base our measures of expected

return and risk on estimates from an observed
sample of realized (historical) returns (i.e. referring to
a number of periods).

(Note that the text prefers to use the term ‘historical’

rather than sample)

We assume the unknown probability distribution of

returns remains approximately the same over the
timeframe of the sample.

If we assume this distribution does not change over time

we can take each sample value as a realization from
this distribution.

9
Clearly if our time series of returns is reasonably long it
is unrealistic to assume that the distribution of returns
has not changed since the period of the first
observation. That is why we do not go far back into the
past when estimating return distribution parameters
from a sample.

Given a sample of n observed returns (R1 ,........, Rn )

1 n
Sample Average(mean )  r   Ri
n i 1

n
1
 Sample  standard deviation of returns  sr   i
( n  1) i 1
( r  r ) 2

(Please note that, we are not following the textbook

notations for calculating risk and return.)

We can use the sample mean and sample standard

deviation as estimators of the mean and standard
deviation of the underlying probability distribution
of returns.
Parameters Symbol
Sample Population
Average 
X  
Standard 
s  
deviation
Correlation 
r  
co-efficient

10
Example:
Suppose we have observed the following realized
returns.

Year Bonds Co. A Co. B

shares shares
2007/08 6.1% 4.2% 3.9%
2008/09 6.4 7.0 6.5
2009/10 5.2 7.2 7.4
2010/11 4.3 8.0 8.2
2011/12 5.0 3.5 6.0
2012/13 6.0 4.3 5.8

Calculate the sample means, standard deviations

and coefficients for these securities:

r Bond  (6.1  6.4  5.2  4.3  5  6.0) 6  5.5

r CompanyA  ( 4.2  7  7.2  8  3.5  4.3) 6  5.7
r CompanyB  (3.9  6.5  7.4  8.2  6  5.8) 6  6.3

11
Standard Deviation (s):

(6.1  5.5)2  (6.4  5.5)2  (5.2  5.5)2  ( 4.3  5.5)2  (5  5.5)2  (6.0  5.5)2
sBond 
5
 0.64  0.8000%
Using same method, we get:
sCompanyA  1.9121%
sCompanyB  1.4805%
Coefficient of variation (CV):
0.8000
CVBond   100%  14.55%
5. 5
1.9121
CVCompanyA   100%  33.54%
5.7
1.4805
CVCompanyB   100%  23.50%
6. 3
s
Please Note that for sample data the CV=  100%
X
NOTE:
These calculations will be easier using scientific
calculator.

Steps to calculate mean and standard deviation:

i) Clear everything from the memory

ii) Go to ‘sd’ mode by pressing the ‘MODE’ button in
the calculator
iii) Enter the data using M+ in the calculator (For fx-
82AU PLUS II only enter)
iv) Enter ‘Shift’ then the appropriate button
(depending on the calculator)

The detail demonstrations will be given during the

lecture. This procedure is most efficient and easy.

12
(All of these calculations can be performed efficiently
using built-in functions on calculators or in Excel)

(c) Attitudes to Risk

Three possible types of attitude to risk by investors are

normally distinguished:

 For a risk averse investor, the required return to

hold an asset is an increasing function of the risk
associated with an asset’s return (i.e. higher risk
implies higher required return-e.g. risk increased
from x1 to x 2 )

 For a risk seeking (risk loving) investor, the required

return to hold an asset is a decreasing function of
the risk associated with an asset’s return (i.e. higher
risk implies lower required return)

 For a risk indifferent (or neutral) investor, the

required return to hold an asset is constant,
regardless of the risk associated with an asset’s
return

13
It is generally assumed that investors are risk
averse.

Return
Risk-averse

Risk-indifferent

Risk-lover
Risk
x1 x2
Risk

14
2. Measuring Expected Return and Risk of a
Portfolio of Assets

In finance, a portfolio is a collection of assets.

Suppose a portfolio consisting of n assets, where

the return on asset j is given by R ( j=1 , .. . ., n ) . If
j

wj represents the proportion of the total dollar

value of the portfolio associated with asset j, then
the return on the portfolio is given by
n
Rp  w1R1  w 2R2  ......  w nRn   w j R j
j 1

(weighted average of returns)

Given that the expected value (mean or average) of a

sum is the sum of the expected values, we can write
the mean return of the portfolio as
n
R p  w1 R1  w2 R 2  ......  wn R n   w j R j
i 1

(In terms of sample or probability distribution

means)

The standard deviation of portfolio returns is not a

simple weighted average of the standard deviations
of the returns of the component assets.

15
For example, for a 2-asset portfolio (of assets 1 and
2) the (population) standard deviation of the
probability distribution of portfolio returns is given
by

 R p  w12 R21  w22 R22  2w1w2 1,2  R1 R2

where:
 ρ1,2 (‘rho’) is the (population) correlation
coefficient of returns on assets 1 and 2. (The
sample or ‘historical’ correlation coefficient is
normally denoted r 1,2 ).
 σ 1 is the standard deviation of returns to asset
1 (analogously for asset 2).

Cov (R 1 ,R 2 )
Population correlation coefficient=1,2 
 R1 R2
Cov (R 1 ,R 2 )
Sample correlation coefficient=r1,2 
sR1 sR2

Where Cov (R1,R 2 ) is the covariance of the returns of

the two assets. If we are given discrete probability
distributions for assets 1 and 2, the covariance
between the two assets’ returns is given by

16
 n
Cov (R1,R 2 )   (R1i  R1 )(R 2 i  R 2 )Pi ; when probability is given
i 1

1 N
Cov (R1,R 2 )   (R1i  R1 )(R 2 i  R 2 ); for population
N i 1
1 n
Cov (R1,R 2 )   (R1i  R1 )(R2i  R 2 ); for sample
n  1 i 1

Where R1i is the return on asset 1 in state of nature i,

etc.

(Students are not expected to remember the above


formulae for R p and Cov (R1,R 2 ) )

Thus the covariance given above is a weighted average

of the products of the deviations from the means of the
returns to the two assets, i.e. over the various states of
nature.

The covariance of returns is a measure of the

degree of linear relationship between the two
returns, i.e. how they co-vary linearly. However the
covariance measure is influenced by the unit of
measure: dividing it by the product of the two
standard deviations removes this influence and
results in a measure (the correlation coefficient) that
can only take values between -1 and +1.

The sample counterparts of the above formulae are

obtained by simply substituting the corresponding
sample estimators.

17
If the correlation coefficient is -1, we have perfect
negative correlation (e.g. between two assets’
returns). If the correlation coefficient is 1, we have
perfect positive correlation. If the correlation
coefficient is 0, there is no linear relationship (even
approximate) between the two assets’ returns
(Illustrate as per diagrams below)

18
Example:

Perfect positive correlation Perfect negative correlation

R2 R2

R1 R1

(We will subsequently say more about correlation)

Given a set of portfolio returns (sample or

probability distribution) we can calculate the
standard deviation of portfolio returns by using the
previous formulae for the standard deviation of
returns to a single asset: thus in this situation we do
not need to calculate the portfolio return standard
deviation using a complicated formula involving the
standard deviations for the returns of the individual
assets in the portfolio.

Example:
Given the previous sample of realized returns,
suppose that half of the value of an investor’s
portfolio is invested in bonds, a quarter in company
A shares and a quarter in company B shares.
Calculate the mean and standard deviation of this
portfolio. (5.75%, 0.6935%)

19
We have (w bonds=0 .5 , w A=0 . 25 , w B =0 . 25)

Year Rp

2007/08 0.5(6.1) + 0.25(4.2) + 0.25(3.9)=5.075

2008/09 6.575
2009/10 6.25
2010/11 6.2
2011/12 4.875
2012/13 5.525

Rp
Sample mean of
 (5.075  6.575  6.25  6.2  4.875  5.525) 6  5.75%
 wBond  rBond  wA  rA  wB  rB
 0.5(5.5)  0.25(5.7)  0.25(6.3)  5.75%

Rp 
Sample standard deviation of
(5 .075−5 .75 )2 +(6 .575−5. 75 )2 +. .. .. .+(5 . 525−5 .75 )2
√ 5
¿ 0 .6935 %

20
 DIVERSIFICATION AND RISK

(a) Correlation and Risk of a Portfolio

The standard deviation of returns is a measure of total

risk or total variability of a portfolio.

The most important aspect of risk is the overall risk

of the firm as viewed by the investors in the market.
Risk affects investment opportunities and investors
wealth.

The total risk of a portfolio can potentially (not

always) be reduced by adding to the portfolio assets
whose returns are less than perfectly positively
correlated with the returns to the existing assets in
the portfolio. Thus diversification can reduce portfolio
risk.

In theory, if the returns on two assets are perfectly

negatively correlated, the assets can be combined in
such a way (i.e. choosing w 1 and w 2 ; w  w  1 ) that
1 2

the standard deviation of the portfolio return would

be minimum, i.e. if ρ1,2 =−1 we can find w 1 and
w 2 such that  Rp  minimum .

Remember that for a 2-asset portfolio

 Rp  w12 R21  w 22 R2 2  2w1w 2 1,2  R1 R2

21
However it is not easy to find many assets whose
returns are perfectly negatively correlated or simply
negatively correlated. Most pairs of asset returns are
positively correlated with a correlation coefficient
between 0.5 and 0.7.

In general, for a 2-asset portfolio:

 If the correlation coefficient between the asset
returns is perfectly negatively correlated, various
Rp 
combinations of the assets will yield a
between 0 (inclusive) and the standard deviation
 most risky  of the most risky asset.
0   Rp   most risky
i.e.
For, 1,2  1
 Rp  w12 R21  w 22 R2 2  2w1w 2 1,2  R1 R2  w12 R21  w 22 R2 2  2w1w 2  R1 R2

w  
2
 Rp  1 R1  w 2  R2  w1 R1  w 2  R2
 If the correlation coefficient is equal to 1, various
 Rp
combinations of the assets will yield a
between the σ ’s of the two assets (inclusive).

For, 1,2  1
 Rp  w12 R21  w 22 R2 2  2w1w 2 1,2  R1 R2  w12 R21  w 22 R2 2  2w1w 2  R1 R2

w  
2
 Rp  1 R1  w 2  R2  w1 R1  w 2  R2
 less risky   Rp   most risky
 If the correlation coefficient is equal to 0, various
 Rp
combinations of the assets will yield a
between the σ ’s of the two assets (inclusive).
22
For, 1,2  0
 Rp  w12 R21  w 22 R2 2  2w1w 2  R1 R2  0

 Rp  w12 R21  w 22 R2 2

 less risky   Rp   most risky

 w1 R1  w 2  R2  w12 R21  w 22 R22  w1 R1  w 2  R2

 Risk (Correlation=-1)<Risk (Correlation=0)<Risk (Correlation=+1)

"Don't put all your eggs to the same basket".

Diversification reduces the risk involved to a portfolio.

Diversification protects the entire portfolio from various
risk and volatility associated to various asset classes. 

Example:
Year Bonds Shares return (%)
return (%)
Weight 60 %( w  0.6 ) 40 %( w  0.4 Portfolio
1 2

)
2007 6.10 3.90 5.22
2008 6.40 6.50 6.44
2009 5.20 7.40 6.08
2010 4.30 8.20 5.86
2011 5.00 6.00 5.40
23
2012 6.00 5.80 5.92
Expected return 5.50 6.30 5.82
Variance 0.64 2.19 0.96
Standard deviation 0.80 1.48 0.98
Correlation coefficient (r) 0.67

Therefore it is clear that diversification will reduce the

risk. This example will be further explained in the
lecture.

Diversifying two risky assets in a portfolio reduce the

total risk. Reduction of the risk will be most effective if
the correlation coefficient between the assets is -1.
(b) Diversifiable (or unsystematic) and Non-
diversifiable (systematic) Risk

The total risk has two components diversifiable risk

and non-diversifiable risk.

Up till now we have only considered total portfolio risk,

measured by  . Rp

It is sometimes useful to divide the total risk of a

security/portfolio into two parts:

 Risk that specifically affects a single or a small group

of (similar) assets is called diversifiable risk.
Diversifiable or un-systematic risk is associated with
random events or firm specific factors (such as
strikes, lawsuits, regulatory actions, loss of key
accounts etc.) that are unrelated to factors affecting
the whole market for risky assets. (For example,
recent fall in iron ore price reduced the share price of
24
 some companies engaged in the extraction of Iron ore
implies a reduction in return.) In theory, diversifiable
risk can be reduced (close to zero) through portfolio
diversification. In general, one can minimize the risk
by adding 20-25 securities (or assets).

 Risk that affects a large number of assets (greater or

lesser degree) called non-diversifiable or systematic
risk. Non-Diversifiable or systematic risk is associated
with factors affecting all assets/securities making up
the market portfolio of risky assets. (for example: war,
economic condition (GNP/GDP, inflation, interest
rate), international incidents, political events, natural
events (flood, famine, earthquake etc.). This type of
risk cannot be minimized by adding more assets in the
portfolio.

Thus total security risk = diversifiable risk + non-

diversifiable risk.

As more and more securities are added to a portfolio,

the level of non-diversifiable risk of the portfolio
approaches that of the market portfolio.

By definition, the market portfolio is a portfolio consisting

of all available risky assets in the same value
proportions as their value proportions of the total value
of all risky assets.

For example, an unanticipated increase in the inflation

affects wage and the price (therefore costs) of the raw
materials of that company buys, the value of the assets
that company own and the price of the finished
25
products. This increase in the unexpected inflation
raises therefore affects most of the companies are an
example of systematic risk. On the other hand a small
oil strike can affect a small group of companies are an
example of unsystematic risk.

Each security/portfolio possesses its own level of

systematic risk.

Textbook Diagram:

Total Portfolio
σ
Risk r p

σ2 Diversifiable risk

Risk associated with

market portfolio
(systematic risk only)
2 Number of securities in portfolio

In the diagram, σ 2 = average total risk associated

with a 2-security portfolio.

The above diagram is rationalized by noting that the

average systematic risk of a 1-security, 2-security,…etc.
portfolio is the same as that of the market portfolio,
although each particular portfolio possesses its own
level of systematic risk.
26
Most benefits of diversification can be achieved with
portfolios consisting of 15-20 securities. Therefore, the
unsystematic or diversifiable risk can be minimized
through diversification, but the systematic risk is not
possible to minimize through diversification. Therefore,
systematic risk is the only relevant risk.

Since any investor can form a portfolio that

achieves the available benefits of diversification,
non-diversifiable risk is the only relevant risk for
investors and valuation purposes.

Rationale for systematic or market risk being the only

relevant risk:
If a security yielded a high enough return to satisfy a
non-diversified investor, it would represent a bargain for
well-diversified investors, who in turn would buy the
security, pushing down its risk premium to that
associated only with its non-diversifiable risk.

Systematic risk also referred as market risk.

27
 Corporate Financial Management

Exercises to be done in the workshop:

It is expected that students will try the problems below at home before attending the
workshop. The lecturer will help to solve these problems during the workshop.
(Be sure to draw timelines when explaining time value of money problems)

1. An individual is to receive $400 a year at the end of each year for the next ten years.
If the rate of interest is 8% p.a. compounded annually, what is the present value of this
annuity?

2. An individual has a water bill for $200 due to be paid today. The water company
charges interest at a rate of 15.6% per annum compounded weekly on overdue
amounts. Although the individual is unable to pay the complete bill today, he accepts
to pay his debt to the water company by making 4 equal weekly payments, with the
first payment to be made today. What should each weekly payment be (assuming
there are 52 weeks in a year)

3. A debt is being paid off with equal payments of $2,000 at the end of each quarter for
two years. If interest on the debt is charged at 8% per annum compounded quarterly,
calculate the principal outstanding (amount owing) when the fourth payment has just
been made.

4. A printing firm borrows $50,000 to purchase a new poster printing machine. Under
the loan agreement, the firm will make 12 equal monthly repayments with the first
repayment made 6 months after the loan amount is transferred to the firm’s bank
account. If the rate of interest charged on the loan is 12% per annum compounded
monthly, what amount will be each of the 12 monthly repayments?

5. A firm borrows $500,000 from a bank. Interest is charged on the loan at a rate of
12% per annum compounded quarterly.

(a) Suppose the firm agrees to pay back the loan by making 40 quarterly
payments over 10 years. Calculate the required quarterly payment if the
first payment is made one quarter after the loan amount is received by the
firm?

(b) Suppose instead that the firm agrees to pay back the loan by making 40
equal quarterly payments with the first payment made one year after the
loan amount is received by the firm. Calculate the quarterly payment in this
case.

6. A company is running a competition in which the first prize is a perpetuity paying

$100 each month. Supposing an interest rate of 9% per annum compounded monthly,
how much money will the company need to fund the prize now if the first payment is
made

(a) in one month's time?

(b) immediately?
(c) a year from now?

28
7. In order to replace a large freezer system in the future, an abattoir plans to deposit 8
equal amounts into a fund at yearly intervals, with the first deposit made immediately,
so that the amount in the fund will be $50,000 in 8 years’ time.
(a) Assuming the fund earns 7% per annum compounded annually, calculate the
annual deposit amount.
(b) Suppose now that although the fund initially earns 7% per annum
compounded annually, after 3 years, i.e. just after the fourth deposit is made,
the interest rate unexpectedly falls to 6% per annum compounded annually. If
the firm changes its annual deposit to take account of the lower interest rate,
what will be the new annual deposit amount for the last 4 deposits?
(Students are strongly advised to draw a timeline when answering this
question, tutor will not be discussing this problem during the tutorial if
time is not available. This question will not be in the final examination)

8. Discussion question: (tutor will not be discussing this problem during the tutorial
this question will not be in the final examination):
A $50,000 loan is to be repaid by 8 quarterly repayments, with the first repayment
made one quarter after the loan amount is received by the borrower. The interest rate
is initially 12% p.a. compounded quarterly. After 12 months, however, the interest
rate unexpectedly decreases to 10% p.a. compounded quarterly. What will be the
quarterly repayments for the second year (i.e. the last 4 quarters), given that the term
of the loan remains unchanged?

29
 MAIN POINTS
 The expected or mean return to an asset is the weighted average of the
possible returns, where the weights are the probabilities of the
particular returns occurring.
 A sample mean return to an asset is an estimate of the expected return.
 Risk associated with the return to an asset is measured by the variance
or standard deviation of returns.
 The ‘population’ standard deviation of returns can be calculated if the
probability distribution of returns is known, whilst the sample standard
deviation of returns is based on a sample of past returns.
 The coefficient of variation of returns allows comparison of the relative
variability of returns to assets with different expected (or sample mean)
returns.
 Investors are normally risk averse.
 The return on a portfolio is the weighted sum of component asset
returns, with each weight given by the proportion of the portfolio’s value
accounted for by the particular asset.
 The correlation coefficient between the returns to two assets represents
the degree of linear relationship between the assets’ returns.
 Systematic risk represents variability of returns due to factors that
affect all risky assets.
 Non-systematic risk represents variability of returns due to factors that
are unrelated to factors affecting all risky assets.
 Portfolio diversification can in theory reduce a portfolio’s non-
systematic risk to zero and its systematic risk to that associated with
the market portfolio.

30