MAC 306:Financial Economics

Topic No.5:Measurement of Investment Risks

Course Lecturer: Dr. Joshua Were

26th July 2018

MASENO UNIVERSITY

[Link]
TOPIC 5
MEASURES OF INVESTMENT RISK
Introduction
Most investment decisions revolve around ’risk’ and return. It is therefore use-
ful to be able to quantify exactly the degree of investment risk that an investor
faces. We outline three such measures, based upon alternative definitions of risk
together with their relative merits.

Measures of Risk
Most mathematical theories of investment use variance of return as a measure
of risk. For example, (mean-variance)t portfolio theory and the capital asset
pricing model, both of which will be discussed later in the course.
However it is not obvious that variances necessarily corresponds to investor’s
perception of risk and other measures have been proposed as being more appro-
priate.
Some investors might be concerned not only with the mean and variance of
returns, but also more generally, with other higher moments of returns, such
as skewness of returns. For example, although two risky assets might yield the
same expectation and variance of future returns,, if the returns on asset A are
positively skewed, whilst those of B are symmetrical about the mean, the asset
A might be preferred to asset B by some investors.

Variance of Return
Variance of return is defined as
Z ∞
(µ − x)2 f (x)dx
−∞

where µ is the mean return at the end of the chosen period and f (x) is the
probability density function of the return.
Variance has the advantage over most other measures in that it is mathemati-
cally tractable and the mean-variance framework leads to elegant solutions for
optimal [Link] use of mean - variance theory has been shown to give a
good approximation to several other proposed methodologies. Mean variance
portfolio theory can be shown to lead to optimum portfolios if investors can
be assumed to have quadratic utility functions or if returns are assumed to be
normally distributed.
If these conditions do not hold then we cannot assume that the investors make
choices wisely on the basis of the mean and variance of return.

Example
Investment returns (%p.a), X, on a particular asset are modelled using a prob-
ability disribution with density function

f (x) = 0.00075(100 − (x − 5)2 ) − 5 ≤ x ≤ 15

Calculate the mean and the variance of return.

Solution
The density function is symmetrical about x = [Link] the mean return is 5%.
Alternatively, this could be found by integrating as follows:
Z 15
E[X] = 0.00075 x(100 − (x − 5)2 )dx
−5
Z 15
= 0.00075 75x + 10x − x3
−5
 15
75 2 10 3 1 4
= 0.00075 x + x − x
2 3 4 −5

= 0.00075 [7031.25 − 364.5833]

=5 i.e 5%
The variance is given by
Z 15
(x − 5)2 100 − (x − 5)2 dx


V ar[X] = 0.00075
−5
Z 15
100(x − 5)2 − (x − 4)4 dx


= 0.00075
−5
 15
100 1
= 0.00075 (x − 5)3 − (x − 5)4
3 3 −5

= 0.00075 [13, 333.33 − (−13, 333.33)]

= 20 i.e 20%

2
Example
Investment return (%p.a), X, on a particular asset are modelled using the prob-
ability distribution

X Probability
-7 0.04
5.5 0.96
Calculate the mean and variance of return.

Solution
The mean and variance is given by

E[X] = −7 × 0.04 + 5.5 × 0.95

=5 i.e 5%
The variance is given by

V ar[X] = (5 − (−7))2 × 0.04 + (5 − 5.5)2 × 0.96

=6 i.e 6%
Alternatively, you may have calculated the variance using the formula
2
V ar[X] = E[X 2 ] − [E[X]]

where E[X 2 ] = 31%%

Semi - Variance of Return

The main argument against the use of variance as a measure of risk is that most
investors do not dislike uncertainty of returns as such rather they dislike the
possibility of low returns.
Presumably all investors would choose a security that that offered a chance of
either a 10% or 12% return in preference to one that offered a certain 10%,
despite the greater uncertainity associated with the former.

One measure that seeks to quantify this view is downside semi-variance.

This is defined as Z µ
(µ − x)2 f (x)dx
−∞

For a discrete random variable, the downside semi-variance is defined as

X
(µ − x)2 P r(X = x)
x<µ

3
Semi-variance is not easy to handle mathematically and it takes no account of
variability above the mean. Furthermore if returns on assets are symmetrically
distributed semi-variance is proportional to variance.
Thus if assets are assumed to be normally distributed then it offers no advan-
tage over the straight forward variance.

Example
Calculate the downside semi-variance for the asset whose probability density
function is given by

f (x) = 0.00075(100 − (x − 5)2 ) − 5 ≤ x ≤ 15

Solution
The continuous distribution is symmetrical. Therefore, the downside semi-
variance is half the variance, i.e 10%%.

Example
Calculate the downside semi-variance of an asset modellled by the following
distribution
X Probability
-7 0.04
5.5 0.96

Solution
The downside semi-variance is given by
X
(5 − x)2 P r(X = x)
x<5

= (5 − (−7))2 × 0.04
= 5.76
i.e
5.76%%

Shortfall Probabilities
A shortfall probability measures the probability of returns below a certain level.
Thus if L is the chosen benchmark level, then the shortfall probability is
Z L
P r(X < L) = f (x)dx
−∞

4
 Alternatively the risk level can be expressed as the expected shortfall below
a certain level. Z L
(L − x)f (x)dx
−∞

Where L is the chosen benchmark level.

The benchmark level can be expressed as the return on a benchmark fund if
this is more appropriate than absolute level. In fact any of the risk measures
discussed can be expressed as measures of the risk relative to a suitable bench-
mark which may be an index, a median fund or some level of inflation.
L could alternatively relate to some pre-specified level of surplus or fund sol-
vency.
Downside risk measures have also been proposed based on an increasing func-
tion of (L − x) rather than (L − x) itself in the integral.

In other words we can use a measure of the form

Z L
g(L − x)f (x)dx
−∞

Two particular cases of note are:

1. g(L − r) = (L − r)2 this is the so called shortfall variance.
2. g(L − r) = (L − r) the average or expected shortfall measure defined above.

Note also that if g(x) = x2 and L = µthen we have the semi - variance
measure defined above.
Shortfall measures are useful in monitoring a fund’s exposure to risk because
the expected under performance relative to a benchmark is a concept that is
apparently easy to understand. As with semi-variance, however, no attention
is paid to the distribution of out performance of the benchmark, i.e returns in
excess of L are again completely ignored.

Example
Calculate the shortfall probability for the asset modelled by the density function
given below, where the benchmark return is 0% pa.

f (x) = 0.00075(100 − (x − 5)2 ) − 5 ≤ x ≤ 15

Solution
The shortfall probability is given by
Z 0
P r(X < 0) = 0.00075 100 − (x − 5)2 dx
5
 0
1
= 0.00075 100x − (x − 5)3
3 5

5
 = 0.00075 [41.6667 − (−166.6667)]
= 0.15625

Example
Calculate the shortfall probability for the asset modelled by the distribution
below, where the benchmark return is 0% pa.

X Probability
-7 0.04
5.5 0.96

Solution
The shortfall probability is given by

P r(X < 0) = 0.04

Value at Risk(VaR)
Value at Risk (VaR) generalises the likelihood of under-performing by providing
a statistical measure of downside risk. For a continuous random variable, Value
at Risk can be determined as:

V aR(X) = −t2cmwhere1cmP (X < t) = p

VaR assesses the potential losses on a portfolio over a given future time period
with a given degree of confidence. For example, if we adopt a 99% confidence
limit, the VaR is the amount of loss that will be exceeded only one time in
a hundred over a given time period and we would need to find t such that
P (X < t) = 0.01. Note that Value at Risk is a ”loss amount”. Therefore:

• a positive Value at Risk (a negative t ) indicates a loss

• a negative Value at Risk (a positive t ) indicates a profit
• Value at Risk should be expressed as a monetary amount and not as a
percentage

Example
Calculate the VaR over one year with a 95% confidence limit for a portfolio
consisting of 100m invested in the asset modelled with the following probability
density function.

f (x) = 0.00075(100 − (x − 5)2 ) − 5 ≤ x ≤ 15

6
Solution
We start by finding t, where P r(X < t) = 0.05
Z t
⇒ 0.00075 100 − (x − 5)2 dx
−5
 t
1
⇒ 0.00075 100x − (x − 5)3 = 0.05
3 −5
Since the equation in brackets is a cubic in t, we are going to need to solve this
equation numerically by trial and error
 −3
1
t = −3 ⇒ 0.00075 100x − (x − 5)3 = 0.028
3 −5
 −2
1 3
t = −2 ⇒ 0.00075 100x − (x − 5) = 0.06075
3 −5
Interpolating between these values gives
0.05 − 0.028
t = −3 +
0.06075 − 0.028
= −2.3
In fact the true value is t = −[Link] t is a percentage investment return
per annum, the 95% value at risk over one year on 100m portfolio is
100m × 2.293% = 2.293m
This means that, we are 95% certain that we will not lose more than 2.293m
over the next year.

Example
Calculate the 95% VaR over one year with a 95% confidence limit for a portfolio
consisting of 100m invested in the asset modelled as follows:
X Probability
-7 0.04
5.5 0.96

Solution
We start by finding t, where t = max {x : P r(X < x) ≤ 0.05} Now P r(X <
−7) = 0 and P r(X < 5.5) = 0.04 .Therefore t = 5.5
Since t is a percentage investment return per annum, the 95% value at risk over
one year on a 100m portfolio is
100 × −5.5% = −5.5m
This means that, we are 95% certain that we will not make profits less than
5.5m over the next year.