Stand alone measures Portfolio measures Required return Summary

Risk and return

The basics
David J. Moore, Ph.D.
www.efcientminds.com
March 12, 2013
Stand alone measures Portfolio measures Required return Summary
Outline
1
Stand alone measures
Denitions and notation
Expected return and risk
Bringing risk and return together
2
Portfolio measures
Portfolio return
Portfolio risk
3
Required return
Total vs. market risk
Calculating required return
4
Summary
What you saw in this presentation...
Stand alone measures Portfolio measures Required return Summary
Denitions and notation
Outline
1
Stand alone measures
Denitions and notation
Expected return and risk
Bringing risk and return together
2
Portfolio measures
Portfolio return
Portfolio risk
3
Required return
Total vs. market risk
Calculating required return
4
Summary
What you saw in this presentation...
Stand alone measures Portfolio measures Required return Summary
Denitions and notation
Historical, expected, required
Historical returns
_
R
_
: The realized or after-the-fact return.
Expected return
_
E [R] =

R
_
: The return that equates the
current price with expected future cash ows.
Required return (R): The return required for delayed
compensation and market risk.
Notation:
Summation
n

i =1

i
=
1
+
2
+ +
n
Product-ation
n

i =1

i
=
1

2

n
Stand alone measures Portfolio measures Required return Summary
Denitions and notation
Arithmetic vs. geometric mean
Arithmetic mean is a forward-looking measure. Statistically it is BLUE (best linear unbiased
estimator). It is the best estimate of next periods value:
arithmetic mean =

R =
1
n
n

t =1
R
t
=
1
n
(R
1
+R
2
+ +R
n
)
Geometric mean is a backward-looking measure. it measures the actual returns
(after-the-fact in IFM10 terminology) realized in the past:
geometric mean = R =
_
n

t =1
RR
t
_
1/n
1 = (RR
1
RR
2
RR
n
)
1/n
1
You must use relative returns RR
t
when computing the geometric
mean.
IFM10 uses R to represent the realized or after the fact return. In
the past I have used R to represent the arithmetic mean and G for
geometric mean. From this point forward I will use

R for arithmetic
mean and R for geometric mean.
Stand alone measures Portfolio measures Required return Summary
Expected return and risk
Outline
1
Stand alone measures
Denitions and notation
Expected return and risk
Bringing risk and return together
2
Portfolio measures
Portfolio return
Portfolio risk
3
Required return
Total vs. market risk
Calculating required return
4
Summary
What you saw in this presentation...
Stand alone measures Portfolio measures Required return Summary
Expected return and risk
Historical vs. probability input data
When dealing with stock returns you are typically given historical
data.
When dealing with sales data you are typically given sales
expectations for various potential states of the world.
The calculations are the same whether you are dealing with
returns, dollar amounts, heights, number of units, etc. Here I
show return examples.
Historical data
Year (t ) R
t
1 10%
2 -5%
3 7%
Probabilities
Demand (i ) Pr
i
R
i
Weak 0.2 -5%
Normal 0.6 7%
Strong 0.2 10%
Stand alone measures Portfolio measures Required return Summary
Expected return and risk
Calculations
Given historical data r
t
Expected return:
E [R] =

R =
1
n
n

t =1
R
t
(1)
The risk (variance) is calculated as:

2
=
1
n1
n

t =1
_
R
t

R
_
2
(2)
Given probabilities Pr
i
and returns R
i
Expected return:
E [R] =

R =
n

i =1
(R
i
Pr
i
) (3)
The risk (variance) is calculated as:

2
=
n

i =1
_
_
R
i

R
_
2
Pr
i
_
(4)
See if you can take the data from the previous slide and compute expected
returns and variances.
Note the standard deviation is the square root of the variance
2
: =

2
Stand alone measures Portfolio measures Required return Summary
Expected return and risk
One more way to calculate expected return
Recall the denition of expected return: The return that equates the current price
with expected future cash ows.
Expressed mathematically:
P
0
=
n

t =1
D
t
(1+R)
t
(5)
If we impose some structure on Eq. (5), specically constant growth of dividends
at rate g, the new equation is:
P =
D
1
Rg
(6)
where D
1
= D
0
(1+g)
We can rearrange Eq. (6) to arrive at another expected return measure:

R =
D
1
P
0
+g =
D
0
P
0
(1+g) +g (7)
In other words, given the current dividend D
0
, the current stock price P
0
, and the
expected dividend growth rate g, we can compute the expected return

R.
Stand alone measures Portfolio measures Required return Summary
Bringing risk and return together
Outline
1
Stand alone measures
Denitions and notation
Expected return and risk
Bringing risk and return together
2
Portfolio measures
Portfolio return
Portfolio risk
3
Required return
Total vs. market risk
Calculating required return
4
Summary
What you saw in this presentation...
Stand alone measures Portfolio measures Required return Summary
Bringing risk and return together
Coefcient of variation
Someone tells you they beat the market with an average
(arithmetic) return of 15% over the past 5 years.
Did they beat the market? The answer is you do not know.
You need to know what level of risk was involved.
Insert coefcient of variation:
CV =

R
CV measures units of total risk per unit of expected return.
Lower CV means less risk and more return lower is better!
Your friend beat the market if they had a lower CV than the
market.
Stand alone measures Portfolio measures Required return Summary
Portfolio return
Outline
1
Stand alone measures
Denitions and notation
Expected return and risk
Bringing risk and return together
2
Portfolio measures
Portfolio return
Portfolio risk
3
Required return
Total vs. market risk
Calculating required return
4
Summary
What you saw in this presentation...
Stand alone measures Portfolio measures Required return Summary
Portfolio return
Portfolio expected return E
_
R
p

=

R
p
You can compute the portfolio expected return two ways:
1
Compute portfolio return by time period t or state of the world i .
Treat R
p,t
or R
p,i
as an individual security. Then apply formulas (1)
and (3) to R
p,t
and R
p,i
, respectively.
2
For each security j compute E
_
R
j

=

R
j
then apply this formula:
E
_
R
p

=

R
p
=
n

j =1
_
w
j

R
j
_
= w
1

R
1
+w
2

R
2
+ +w
n

R
n
(8)
Presuming 40% in stock A (w
a
= 0.40) and 60% in stock B (w
b
= 0.60):
Historical data
Year (t ) R
a,t
R
b,t
R
p,t
1 10% 3% 5.8%
2 -5% 4% 0.4%
3 7% 6% 6.4%
Probabilities
Demand (i ) Pr
i
R
a,i
R
b,i
R
p,i
Weak 0.2 -5% 3% -0.2%
Normal 0.6 7% 4% 5.2%
Strong 0.2 10% 6% 7.6%
Stand alone measures Portfolio measures Required return Summary
Portfolio risk
Outline
1
Stand alone measures
Denitions and notation
Expected return and risk
Bringing risk and return together
2
Portfolio measures
Portfolio return
Portfolio risk
3
Required return
Total vs. market risk
Calculating required return
4
Summary
What you saw in this presentation...
Stand alone measures Portfolio measures Required return Summary
Portfolio risk
Portfolio risk
Unlike portfolio expected return, portfolio risk is not the weighted average of the individual
variances:
2
p
=w
i

2
i
.
In reality, the portfolio risk will be less. A brief look at portfolio theory is provided in later
slides.
For now, we have enough data and tools to compute portfolio risk without worrying about
portfolio theory.
Begin with portfolio return by time period R
p,t
or state of the world R
p,i
and the portfolio
expected return

R
p
.
Given historical data R
t

2
p
=
1
n1
n

t =1
_
R
p,t

R
p
_
2
(9)
Given probabilities Pr
i
and returns R
i

2
p
=
n

i =1
_
_
R
p,i

R
p
_
2
Pr
i
_
(10)
Stand alone measures Portfolio measures Required return Summary
Portfolio risk
Portfolio theory: co-movement
The absolute measure of co-movement is covariance
ab
.
Given historical data R
t

ab
=
1
n1
n

t =1
_
R
a,t

R
a
__
R
b,t

R
b
_
(11)
Given probabilities Pr
i
and returns R
ai
and R
bi

ab
=
n

i =1
_
R
a,i

R
a
__
R
b,i

R
b
_
Pr
i
(12)
The relative relationship between co-movements of returns of returns is the correlation
coefcient :
=
_

_
+1.0 perfect positive correlation
0 no correlation
1.0 perfect negative correlation
Covariance and correlation are related in the following manner:

ab
=

ab

b
(13)
Stand alone measures Portfolio measures Required return Summary
Portfolio risk
Portfolio theory: portfolio risk
For the two-security case portfolio risk
p
is calculated as:

2
p
= w
2
a

2
a
+w
2
b

2
b
+2w
a
w
b

ab

b
(14)
In general for n securities:

2
p
=
n

i =1
w
2
i

2
i
+
n

i =1
n

j =1
i =j
w
i
w
j

ij
(15)
=
n

i =1
n

j =1
w
i
w
j

ij
(16)
=
n

i =1
n

j =1
w
i
w
j

ij

j
(17)
Note: as the number of securities increases, the importance of each securitys
variance decreases.
For fun, compute the portfolio risk using Eqs. (9) and (14).
Stand alone measures Portfolio measures Required return Summary
Total vs. market risk
Outline
1
Stand alone measures
Denitions and notation
Expected return and risk
Bringing risk and return together
2
Portfolio measures
Portfolio return
Portfolio risk
3
Required return
Total vs. market risk
Calculating required return
4
Summary
What you saw in this presentation...
Stand alone measures Portfolio measures Required return Summary
Total vs. market risk
Total and market risk
The total risk measure has two components
Diversiable or idiosyncratic risk that we do not bother measuring.
Non-diversiable or market risk measured by .
An investor is not rewarded for bearing diversiable risk. Why
should you receive additional return per unit of risk that you could
have eliminated?
You are compensated for bearing non-diversiable market risk.
For more information see IFM10 page 52.
Stand alone measures Portfolio measures Required return Summary
Total vs. market risk
Measuring market risk
Market risk is measured by .
Beta is obtained by performing the market model regression:
R
t
= +R
m,t
Where R
t
is the return of the stock in question at time t and R
m,t
is the return of the market at time t .
Interpretation
<1 Below average risk
= 1 Average risk
>1 Above average risk
Calculation of beta in a portfolio is straightforward:

p
=
n

j =1
w
j

j
Stand alone measures Portfolio measures Required return Summary
Calculating required return
Outline
1
Stand alone measures
Denitions and notation
Expected return and risk
Bringing risk and return together
2
Portfolio measures
Portfolio return
Portfolio risk
3
Required return
Total vs. market risk
Calculating required return
4
Summary
What you saw in this presentation...
Stand alone measures Portfolio measures Required return Summary
Calculating required return
From market risk to required return
When you purchase a stock you are (1) delaying your
consumption and (2) taking on market risk. You must be
compensated for both.
The total compensation is called the required return.
Required return R is measured by the Security Market Line
(SML) or CAPM equation:
R = R
f
+ (E [R
m
] R
f
)
where
R
f
= compensation for delayed consumption
= number of units of market risk
E [R
m
] R
f
= market risk premium
= compensation for one unit of market risk
(E [R
m
] R
f
) = compensation for bearing market risk
Stand alone measures Portfolio measures Required return Summary
What you saw in this presentation...
Outline
1
Stand alone measures
Denitions and notation
Expected return and risk
Bringing risk and return together
2
Portfolio measures
Portfolio return
Portfolio risk
3
Required return
Total vs. market risk
Calculating required return
4
Summary
What you saw in this presentation...
Stand alone measures Portfolio measures Required return Summary
What you saw in this presentation...
Many things...
The distinction between historical return
_
R
_
, expected return
_

R
_
, and required return (R).
How to calculate historical return with the geometric mean: R
How to calculate expected return given historical data (arithmetic
mean), when given a probability table, and with current market
information (
D
1/P
0
, g):

R
How to calculate risk (variance): variance
2
for individual
securities; covariance
ab
and
ab
between securities.
How to calculate portfolio expected return

R
p
and portfolio risk

2
p
two ways.
How to calculate required return using CAPM:
R
i
= R
f
+ (E [R
m
] R
f
)
Note: this applies to individual assets and portfolios.