INTRODUCTION TO

VALUE AT RISK (VaR)

ALAN ANDERSON, Ph.D.

ECI Risk Training
www.ecirisktraining.com
Value at Risk (VaR) is a statistical
technique designed to measure the
maximum loss that a portfolio of assets
could suffer over a given time horizon
with a specified level of confidence

(c) ECI Risk Training

www.ecirisktraining.com 2
Value at Risk was originally used to
measure market risk

It has since been extended to other

types of risk, such as credit risk and
operational risk

(c) ECI Risk Training

www.ecirisktraining.com 3
EXAMPLE

Suppose that it is determined that a

$100 million portfolio could potentially
lose $20 million (or more) once every
20 trading days

(c) ECI Risk Training

www.ecirisktraining.com 4
The VaR of this portfolio equals $20
million with a 95% level of confidence
over the coming trading day; 19 out of
20 trading days (95% of the time),
losses are less than $20 million

(c) ECI Risk Training

www.ecirisktraining.com 5
At the 95% confidence level, VaR represents
the border of the 5% “left tail” of the normal
distribution, also known as the fifth percentile
or .05 quantile of the normal distribution

(c) ECI Risk Training

www.ecirisktraining.com 6
 (c) ECI Risk Training
www.ecirisktraining.com 7
This diagram shows that:

95% of the time, the portfolio’s

value remains above $80 million

5% of the time, the portfolio’s

value falls to $80 million or less
(c) ECI Risk Training
www.ecirisktraining.com 8
The VaR of this portfolio is therefore

$100 million - $80 million = $20 million

(c) ECI Risk Training

www.ecirisktraining.com 9
VaR is based on the assumption that the
rates of return of the assets held in a
portfolio are jointly normally distributed

(c) ECI Risk Training

www.ecirisktraining.com 10
VaR has the advantage that the risks
of different assets can be combined to
produce a single number that reflects
the risk of a portfolio

(c) ECI Risk Training

www.ecirisktraining.com 11
Further, the probability of a given
loss can be calculated using VaR

VaR can also be used to determine

the impact on risk of changes in a
portfolio’s composition

(c) ECI Risk Training

www.ecirisktraining.com 12
VaR has the disadvantage that it
is computationally intensive and
requires major adjustments for
non-linear assets, such as options

(c) ECI Risk Training

www.ecirisktraining.com 13
COMPUTING VaR

Value-at-Risk is based on the work of

Harry Markowitz, who was awarded
the Nobel Prize in Economics in 1990
for his pioneering research in the area
of portfolio theory

(c) ECI Risk Training

www.ecirisktraining.com 14
Portfolio theory shows how
risk can be reduced by holding
a well-diversified set of assets

(c) ECI Risk Training

www.ecirisktraining.com 15
A collection of assets is considered to be well-
diversified if the assets are affected differently
by changes in economic variables, such as
interest rates, exchange rates, etc.

(c) ECI Risk Training

www.ecirisktraining.com 16
As a result, a well-diversified portfolio is
less likely to experience extreme changes
in value; in this way, risk is reduced

(c) ECI Risk Training

www.ecirisktraining.com 17
In statistical terms, a well-diversified portfolio
contains assets whose rates of return have
very low or negative correlations with each
other

(c) ECI Risk Training

www.ecirisktraining.com 18
 EXAMPLE

A portfolio consisting exclusively of oil

stocks would not be well-diversified, since
changes in the price of oil would have a
huge impact on the portfolio’s value

(c) ECI Risk Training

www.ecirisktraining.com 19
A portfolio invested in both oil stocks
and automotive stocks would be far
more diversified:

(c) ECI Risk Training

www.ecirisktraining.com 20

Rising oil prices would hurt the automotive
stocks while helping the oil stocks

Falling oil prices would hurt the oil stocks

while helping the automotive stocks

(c) ECI Risk Training

www.ecirisktraining.com 21
As a result, the impact of oil price
swings would be offset by changes in
the value of the automotive stocks

On balance, risk would be reduced

(c) ECI Risk Training

www.ecirisktraining.com 22
The risk of holding a portfolio containing two
assets, X and Y, is measured by its standard
deviation, as follows:

(c) ECI Risk Training

www.ecirisktraining.com 23
 P = w  + w  + 2wX wY
 X  Y
2
X
2
X
2
Y
2
Y

(c) ECI Risk Training

www.ecirisktraining.com 24
where:

P = the standard deviation

of the returns to the portfolio

(c) ECI Risk Training

www.ecirisktraining.com 25

X = standard deviation of
the returns to asset X

Y = standard deviation of
the returns to asset Y

(c) ECI Risk Training

www.ecirisktraining.com 26
wX = weight of asset X
wY = weight of asset Y

The weights represent the proportion

of the portfolio invested in each asset;
the sum of the weights is one

(c) ECI Risk Training

www.ecirisktraining.com 27
NOTE

If short-selling is not possible, then:

0
wX
1
0
wY
1

If short-selling is possible, the

weights can be negative
(c) ECI Risk Training
www.ecirisktraining.com 28
 = “rho”

this represents the correlation

between the returns to assets
X and Y; -1

1

(c) ECI Risk Training

www.ecirisktraining.com 29
The lower is the correlation
between assets, the lower will
be the risk of the portfolio

(c) ECI Risk Training

www.ecirisktraining.com 30
The Value at Risk of a
portfolio is a function of:

(c) ECI Risk Training

www.ecirisktraining.com 31

the dollar value of the portfolio

the portfolio standard deviation

the confidence level

the time horizon

(c) ECI Risk Training

www.ecirisktraining.com 32
COMPUTING VaR FOR
A SINGLE ASSET

For a single asset, using daily

returns data at a confidence level
of c, the VaR is computed as:

V0

(c) ECI Risk Training
www.ecirisktraining.com 33
where:

V0 = initial value of the asset

= standard deviation of the

asset’s daily returns

(c) ECI Risk Training

www.ecirisktraining.com 34

= the number of standard deviations
below the mean corresponding to
the (1-c) quantile of the standard
normal distribution

(c) ECI Risk Training

www.ecirisktraining.com 35
EXAMPLE
For a 95% confidence level, c = 0.95

(1-c) is the fifth quantile (1-.95 = .05 =

5%) of the standard normal distribution

The corresponding value of
is 1.645

(c) ECI Risk Training

www.ecirisktraining.com 36
 (c) ECI Risk Training
www.ecirisktraining.com 37
The value of
corresponding to any
confidence level can be found with a
normal table or with the Excel function
NORMSINV

(c) ECI Risk Training

www.ecirisktraining.com 38
 EXAMPLE

For a 99% confidence level, the value

of
can be determined as follows:

(c) ECI Risk Training

www.ecirisktraining.com 39
c = 0.99
(1-c) = 0.01 = 1%
NORMSINV(0.01) = -2.33

= 2.33

(c) ECI Risk Training

www.ecirisktraining.com 40
 (c) ECI Risk Training
www.ecirisktraining.com 41
 EXAMPLE

Suppose that an investor’s portfolio consists

entirely of $10,000 worth of IBM stock.

Since the portfolio only contains IBM stock,

it can be thought of as a single asset

(c) ECI Risk Training

www.ecirisktraining.com 42
Assume that the standard deviation of the
stock’s returns are 0.0189 (1.89%) per day

(c) ECI Risk Training

www.ecirisktraining.com 43
If the investor wants to know his
portfolio’s VaR over the coming
trading day at the 95% confidence
level, this would be calculated as
follows:

(c) ECI Risk Training

www.ecirisktraining.com 44
V0
 = (10,000)(1.645)(0.0189)

= $310.905

(c) ECI Risk Training

www.ecirisktraining.com 45
This means that over the coming day,
there is a 5% chance that the investor’s
losses could reach $310.905 or more
(i.e., the portfolio’s value could fall to
$9,689.095 or less)

(c) ECI Risk Training

www.ecirisktraining.com 46
NOTE
VaR can be extended to different
time horizons by applying the square
root of time rule

(c) ECI Risk Training

www.ecirisktraining.com 47
According to this rule, the standard
deviation increases in proportion to
the square root of time:

 t
periods = t  1
period

(c) ECI Risk Training

www.ecirisktraining.com 48
If the investor wants to know his
portfolio’s VaR over the coming
month at the 95% confidence level,
based on the assumption that there
are 22 trading days in a month, this
would be calculated as follows:

(c) ECI Risk Training

www.ecirisktraining.com 49
 V0
 = (10, 000)(1.645)(0.0189 22)

(10, 000)(1.645)(0.0189 22) = $1, 458.27

(c) ECI Risk Training

www.ecirisktraining.com 50
Similarly, if the investor wants to know
what his portfolio’s VaR is over the coming
year, assuming that there are 252 trading
days in a year, the calculations would be:

(c) ECI Risk Training

www.ecirisktraining.com 51
 V0
 = (10, 000)(1.645)(0.0189 252)

(10, 000)(1.645)(0.0189 252) = $4,935.46

(c) ECI Risk Training

www.ecirisktraining.com 52
COMPUTING PORTFOLIO VaR

In order to compute the Value at

Risk of a portfolio of two or more
assets, the correlations among the
assets must be explicitly considered

The lower these correlations, the

lower will be the resulting VaR

(c) ECI Risk Training

www.ecirisktraining.com 53
 The Value at Risk of a portfolio
is calculated by determining the:

weight (proportion of the total

invested) of each asset in the
portfolio

(c) ECI Risk Training

www.ecirisktraining.com 54

standard deviation of each asset’s
rate of return in the portfolio

correlations among the assets’ rates

of return in the portfolio

(c) ECI Risk Training

www.ecirisktraining.com 55
Once a confidence level and a time
horizon have been chosen, the
weights, volatilities and correlations
can be combined using Markowitz’s
approach to derive the portfolio’s VaR

(c) ECI Risk Training

www.ecirisktraining.com 56
EXAMPLE
Assume that a $100,000 portfolio
contains $60,000 worth of Stock X
and $40,000 worth of Stock Y.

(c) ECI Risk Training

www.ecirisktraining.com 57
Given the following data, compute
the VaR of this portfolio with a 95%
confidence level over the coming:

(c) ECI Risk Training

www.ecirisktraining.com 58

day

month

year

(c) ECI Risk Training

www.ecirisktraining.com 59
DATA
wX = 0.60 wY = 0.40
X = 0.016284 Y = 0.015380

= -0.19055

(c) ECI Risk Training

www.ecirisktraining.com 60

P = (0.6) (0.016284) + (0.4) (0.015380) +
2 2 2 2

2(0.6)(0.4)(
0.19055)(0.016284)(0.015380)

= 0.01144627 = 1.144627%

(c) ECI Risk Training

www.ecirisktraining.com 61
The portfolio VaR over the coming day is:

V0
 P = (100,000)(1.645)(0.01144627)

= $1,882.91

(c) ECI Risk Training

www.ecirisktraining.com 62
The portfolio VaR over the coming month is:

V0
 P = (100, 000)(1.645)(0.01144627 22)

= $8,831.638

(c) ECI Risk Training

www.ecirisktraining.com 63
 The portfolio VaR over the coming year is:

V0
 P = (100, 000)(1.645)(0.01144627 252)

= $29,890.29

(c) ECI Risk Training

www.ecirisktraining.com 64