MEASURES OF RISK AND

PERFORMANCE
1
MEASURES OF RISK
 Standard deviation of returns, also known as volatility, is the
most common measure of total financial risk.
 If the return distribution is a well-known distribution such as
the normal distribution, then the standard deviation reveals
much of or even all of the information about the width of the
distribution.
 If the distribution is not well-known, then standard deviation is
usually a first pass at describing the dispersion.
 However, standard deviation can be an ineffective measure of
risk when a distribution is non-symmetrical.
 Standard deviation incorporates dispersion from both the right-
hand side (typically profit) and the left-hand side (typically
loss) of the distribution.
 The two sides are identical in a symmetrical distribution, but in
a non-symmetrical distribution the sides differ; and in the case
of risk, the analyst is primarily concerned with the left, or 2
downside, half of the distribution.
MEASURES OF RISK
 Semi-variance
 Semistandard Deviation

 Shortfall Risk

 Target Semivariance

 Target Semistandard Deviation

 Tracking Error

 Drawdown

 Value at Risk

3
SEMI-VARIANCE
 Variance, as a symmetrical calculation, is an
expected value of the squared deviations,
including both negative and positive
deviations.
 The semi-variance uses a formula

otherwise identical to the variance formula

except that it considers only the negative
deviations.
 Semi-variance is therefore expressed as:

4
SEMI-VARIANCE
 Semi-variance provides a sense of how much
variability exists among losses or, more
precisely, among lower-than-expected
outcomes.
 The equation for the semi-variance of a

sample is given as:

5
SEMI-STANDARD DEVIATION
 Semistandard deviation, sometimes called
semideviation, is the square root of
semivariance.
 Most statisticians define T in the computation

of the semivariance and semistandard

deviation as the total number of observations
for a series.
 Some practitioners define T as the number of

observations that have a negative deviation.

 Defining T as including all observations has

desirable statistical properties and is the

6
standard in statistics.
SHORTFALL RISK
 In addition to measuring return risk relative
to a mean return or an expected return,
some analysts measure risk relative to a
target rate of return (such as 5%), chosen by
the investor based on the investor’s goals
and financial situation.
 Generally, the target return is a constant.

 Shortfall risk is simply the probability that

the return will be less than the investor’s

target rate of return.

7
TARGET SEMIVARIANCE AND TARGET
SEMISTANDARD DEVIATION
 The concept of a target return can also be
used in measures of downside dispersion.
 Target semivariance is similar to

semivariance except that target semivariance

substitutes the investor’s target rate of return
in place of the mean return.
 Thus, target semivariance is the dispersion of

all outcomes below some target level of

return rather than below the sample mean
return.
 Target semistandard deviation (TSSD) is

simply the square root of the target 8

semivariance.
TRACKING ERROR
 Tracking error indicates the dispersion of
the returns of an investment relative to a
benchmark return, where a benchmark
return is the contemporaneous realized
return on an index or peer group of
comparable risk.
 Tracking error is usually defined as the

standard deviation of those deviations.

9
DRAWDOWN
 Drawdown is defined as the loss in the
value of an asset over a specified time
interval and is usually expressed in
percentage-return form rather than currency.
 For example, an asset reaching a high of

$100 and then falling to a subsequent low of

$60 would be said to have suffered a
drawdown of 40%.
 Maximum drawdown is defined as the

largest decline over any time interval within

the entire observation period.
10
VALUE AT RISK
 VaR is the maximum loss over a specified
time period within a specified probability.
 The specification of a VaR requires two

parameters:
 The length of time involved in measuring the
potential loss
 The probability used to specify the confidence
that the given loss figure will not be exceeded

11
VALUE AT RISK
 Thus, we might estimate the VaR for a 10-day period
with 99% confidence as being $100,000.
 In this case, the VaR is a prediction that over a 10-
day period, there is a 99% chance that performance
will be better than the scenario in which there is a
$100,000 loss.
 Conversely, there is a 1% chance that there will be a
loss in excess of $100,000,
 Conditional value-at risk (CVaR), also known as
expected tail loss, is the expected loss of the investor
given that the VaR has been equaled or exceeded.
 Thus, if the VaR is $1 million, then the CVaR would be
the expected value of all losses equal to or greater
12
than $1 million.
ESTIMATING VAR WITH NORMALLY
DISTRIBUTED RETURNS
 A VaR computation assuming normality and
using the statistics of the normal distribution is
known as parametric VaR.

N is the number of standard deviations, which

depends on the confidence level that is specified.
 σ is the estimated daily standard deviation
expressed as a proportion of price or value (return
standard deviation).
 Square root of the number of days used for the
VaR analysis, such as 1, 2, 5, or 10 days.
 Value is the market value of the position for which 13

the VaR is being computed.

ESTIMATING VAR WITH NORMALLY
DISTRIBUTED RETURNS
 Let’s return to the example of JAC Fund’s $1
million holding of the ETF with an expected
return of zero. Estimating roughly that the
daily standard deviation of the ETF is 1.35%,
for a 99% confidence interval, Find the 10-
day VaR?

14
TWO PRIMARY APPROACHES TO
ESTIMATING THE VOLATILITY FOR VAR
 A common approach is to estimate the standard
deviation as being equal to the asset’s historical
standard deviation of returns.
 Much work has been done and is being devoted

to developing improved forecasts of volatility

using past data.
 These efforts focus on the extent to which more

recent returns should be given a higher weight

than returns from many time periods ago.
 Models such as ARCH and GARCH emphasize

more recent observations in estimating volatility

based on past data. 15
TWO PRIMARY APPROACHES TO
ESTIMATING THE VOLATILITY FOR VAR
 Another method of forecasting volatility is
based on market prices of options.
 Estimates of volatility are based on the

implied volatilities from option prices.

 These estimates, when available and

practical, are typically more accurate than

estimates based on past data, since they
reflect expectations of the future.

16
TWO APPROACHES TO ESTIMATING VAR FOR
LEPTOKURTIC POSITIONS
 The VaR computations are sensitive to mis-
estimation of the probabilities of highly
unusual events.
 If a position’s risk is well described by the

normal distribution, then the probabilities of

extreme events are easily determined using
an estimate of volatility.
 But leptokurtic positions have fatter tails

than the normal distribution, so VaR is

sensitive to the degree to which the
position’s actual tails exceed the tails of a
normal distribution. 17
TWO APPROACHES TO ESTIMATING VAR FOR
LEPTOKURTIC POSITIONS
 One solution is to use a probability
distribution that allows for fatter tails.
 For example, the t-distribution not only

allows for fatter tails than the normal

distribution but also has a parameter that
can be adjusted to alter the fatness of the
tails.
 A second and potentially simpler approach to

adjusting for fat tails is simply to increase the

number of standard deviations in the formula
for a given confidence level.
18
ESTIMATING VAR DIRECTLY FROM
HISTORICAL DATA
 A very simple way to estimate VaR can be to
view a large collection of previous price
changes and compute the size of the price
change for which the specified percentage of
outcomes was lower.
 For example, consider a data set with a long-

term history of deviations of an ETF’s return

from its mean return. We wish to estimate a
five-day 99% VaR.

19
ESTIMATING VAR DIRECTLY FROM
HISTORICAL DATA
 We might collect the daily percentage price
changes of the ETF for the past 5,000 days and
use them to form 1,000 periods of five days each.
 We then rank the five-day deviations from the
highest to the lowest.
 Suppose we find that exactly 10 of these 1,000
periods had price drops of more than 6.8%, and
all the rest of the periods (99%) had better
performance.
 The 99% five-day VaR for our ETF position could
then be estimated at 6.8% of the portfolio’s
current value, under the assumption that past
price changes are representative of future price 20

changes.
ESTIMATING VAR WITH MONTE CARLO
ANALYSIS
 Monte Carlo analysis is a type of
simulation in which many potential paths of
the future are projected using an assumed
model, the results of which are analyzed as
an approximation to the future probability
distributions.
 In finance, it can be very complex to use a

model to solve directly for the probability of a

given loss in a complex portfolio that
experiences a variety of market events, such
as interest rate shifts.
21
ESTIMATING VAR WITH MONTE CARLO
ANALYSIS
 To address the problem with Monte Carlo simulation,
the risk manager defines how the market parameters,
such as interest rates, might behave over the future
and then programs a computer to project thousands
and thousands of possible scenarios of interest rate
changes and other market outcomes.
 Each scenario is then used to estimate results in
terms of the financial outcomes on the portfolio being
analyzed.
 These results are then used to form a probability
distribution of value changes and estimate a VaR.
 A Monte Carlo simulation might project one million
outcomes to a portfolio. In that case, a 99% VaR would
be the loss that occurred in the 10,000th worst 22
outcome.
THREE SCENARIOS FOR
AGGREGATING VAR
 The fund’s analyst reports a VaR of $100,000
for position #1 and a VaR of $100,000 for
position #2. The critical question is the VaR
of the combined positions.
 Let’s consider three scenarios based on

correlations between the returns of the two

positions:

23
THREE SCENARIOS FOR
AGGREGATING VAR
 Perfect positive correlation: If the two positions
are identical or have perfectly positive correlated
and identical risk exposures, then the VaR of the
combination is simply the sum of the individual
VaRs, $200,000.
 Zero correlation: If the two positions have
statistically independent risk exposures, then under
some assumptions, such as normally distributed
outcomes, the VaR of the combination might be the
sum of the individual VaRs divided by the square
root of 2, or $141,421.
 Perfect negative correlation: If the two positions
completely hedge each other’s risk exposures, then
24
the VaR of the combination would be $0.
RATIO-BASED PERFORMANCE
MEASURES
 The Sharpe Ratio
 The Treynor Ratio

 The Sortino Ratio

 The Information Ratio

 Return on VaR

25
THE SHARPE RATIO
 Consider a portfolio that earns 10% per year
and has an annual standard deviation of 20%
when the risk-free rate is 3%.

26
THE TREYNOR RATIO
 Consider a portfolio that earns 10% per year
and has a beta with respect to the market
portfolio of 1.5 when the risk-free rate is 3%.

27
THE SORTINO RATIO
 Consider a portfolio that earns 10% per year
when the investor’s target rate of return is
8% per year. The semistandard deviation
based on returns relative to the target is 16%
annualized.

28
THE INFORMATION RATIO
 If a portfolio consistently outperformed its
benchmark by 4% per year, but its
performance relative to that benchmark
typically deviated from that 4% mean with an
annualized standard deviation of 10%.

29
RETURN ON VAR
 Return on VaR (RoVaR) is simply the
expected or average return of an asset
divided by a specified VaR (expressing VaR as
a positive number):

30
RISK-ADJUSTED RETURN MEASURES
 Jensen’s Alpha
 M2 (M-Squared) Approach

31
JENSEN’S ALPHA
 A portfolio is expected to earn 7% annualized
return when the riskless rate is 4% and the
expected return of the market is 8%. If the
beta of the portfolio is 0.5. Find Jensen’s
Alpha for the portfolio?

32
M2 (M-SQUARED) APPROACH
 The M2 approach, or M-squared approach, expresses
the excess return of an investment after its risk has
been normalized to equal the risk of the market
portfolio.
 The first step is to leverage or deleverage the
investment so that its risk matches the risk of the
market portfolio.
 The superior return that the investment offers relative
to the market when it has been leveraged or
deleveraged to have the same volatility as the market
portfolio is M2.
 A fund is leveraged to a higher level of risk when money
is borrowed at the riskless rate and invested in the fund,
and a fund is deleveraged when money is allocated to
33
the riskless asset rather than invested in the fund.
M2 (M-SQUARED) APPROACH

34
M2 (M-SQUARED) APPROACH
 Consider a portfolio with M2 = 4%. The
portfolio is expected to earn 10%, while the
riskless rate is only 2%. What is the ratio of
the volatility of the market to the volatility of
the portfolio?

35