Value at Risk

Dr. Handayani, [Link], MM, MHP, HIA, FLMI, AFSI, AAK, AAIJ, AMRP, FSAI
Definition of VaR
When using the value at risk measure, we are interested in making a
statement of the following form:

“We are X percent certain that we will not lose more than V dollars in time T.”

The variable V is the VaR of the portfolio. It is a function of two parameters:

the time horizon, T, and the confidence level, X percent. It is the loss level
during a time period of length T that we are X% certain will not be exceeded.
Definition of VaR
VaR can be calculated from either the probability distribution of gains during
time T or the probability distribution of losses during time T (in the former
case, losses are negative gains; in the latter case, gains are negative losses.).
For example, when T is five days and X = 97, VaR is the loss at the 3rd
percentile of the distribution of gains over the next five days. Alternatively, it
is the loss at the 97th percentile of the distribution of losses over the next five
days. More generally, when the distribution of gains is used, VaR is equal to
minus the gain at the (100 − X) percentile of the distribution.
Definition of VaR
The standard deviation of a distribution, although a useful measure in
many situations, does not describe the tails of a probability distribution.
VaR is an important risk measure that focuses on adverse events and
their probability.
The VaR for an investment opportunity is a function of two parameters:
• The time horizon, and
• The confidence level.
Definition of VaR
VaR is the loss level that we do not expect to be exceeded over the time
horizon at the specified confidence level. In this chapter we are
concerned with the distribution of the dollar amount of losses. Gains
are counted as negative losses.
Suppose that the time horizon is ten days, and the confidence level is
99%. A VaR of 10 million would mean that we are 99% certain that the
loss during the next ten days will be less than 10 million. To put this
another way, there is a probability of only 1 % that the loss over the next
ten days will be greater than 10 million.
Definition of VaR
Suppose that the probability density function for the loss during the
time horizon is as shown below. The VaR with a confidence level of 𝑋%
is found by searching for the loss that has an (100 − 𝑋)% chance of
being exceeded. As indicated in the figure, it is the value such that the
area under the distribution equals (100 − 𝑋)%.
Definition of VaR
Suppose that the gain from a portfolio during six months is normally
distributed with a mean of 2 million and a standard deviation of 10
million. From the properties of the normal distribution, the one-
percentile point of this distribution is 2 + (−2.326 × 10) or –21.3 million.
The VaR for the portfolio with a time horizon of six months and a
confidence level of 99% is therefore 21.3 million.
Example 1
Suppose that the gain from a portfolio during six months is normally
distributed with a mean of 2 million and a standard deviation of 10
million. Find the VaR for the portfolio with a time horizon of six months
and confidence level of 99% .
Solution 1
From the properties of the normal distribution, the one-percentile point of
this distribution is 2 + (−2.326348 × 10) or –21.27 million.
The VaR for the portfolio with a time horizon of six months and confidence
level of 99% is therefore 21.27 million.

Notes:
𝑉𝑎𝑟𝛼% = 𝑧1−𝛼 𝜎
𝜇 + (𝑧1% × 𝜎)

Z score can be calculated using function [=[Link]()] in Excel.

Example 2
Suppose that the loss from a portfolio during six months is normally
distributed with a mean of 2 million and a standard deviation of 10
million. Find the VaR for the portfolio with a time horizon of six months
and confidence level of 99% .
Solution 2
From the properties of the normal distribution, the one-percentile point of
this distribution is -2 + (2.326348 × 10) or 21.27 million.
The VaR for the portfolio with a time horizon of six months and confidence
level of 99% is therefore 21.27 million.

Notes:
𝑉𝑎𝑟𝛼% = 𝑧𝛼 𝜎
𝜇 + (𝑧99% × 𝜎)

Z score can be calculated using function [=[Link]()] in Excel.

Single-Asset Case
We now consider how VaR is calculated using the model-building
approach in a very simple situation where the portfolio consists of a
position in a single stock. The portfolio we consider is one consisting of
$10 million in shares of Microsoft. We suppose that 𝑁 = 10 and 𝑋 =
99, so that we are interested in the loss level over 10 days that we are
99% confident will not be exceeded. Initially, we consider a one-day
time horizon.
Single-Asset Case
Assume that the volatility of Microsoft is 2% per day (corresponding to
about 32% per year). Because the size of the position is $10 million, the
standard deviation of daily changes in the value of the position is 2% of
$10 million, or $200,000. It is customary in the model-building approach
to assume that the expected change in a market variable over the time
period considered is zero. This is not exactly true, but it is a reasonable
assumption.
Single-Asset Case
The expected change in the price of a market variable over a short time
period is generally small when compared with the standard deviation of
the change. Suppose, for example, that Microsoft has an expected
return of 20% per annum. Over a one-day period, the expected return is
0.20/252, or about 0.08%, much less than the 2% standard deviation of
the return. Over a 10-day period, the expected return is 0.08 × 10, or
about 0.8%, whereas the standard deviation of the return is 2 10, or
about 6.3%.
Single-Asset Case
So far, we have established that the change in the value of the portfolio
of Microsoft shares over a one-day period has a standard deviation of
$200,000 and (at least approximately) a mean of zero. We assume that
the change is normally distributed. From the Excel [Link]
function, 𝑁(−2.326) = 0.01. This means that there is a 1% probability
that a normally distributed variable will decrease in value by more than
2.326 standard deviations. Equivalently, it means that we are 99%
certain that a normally distributed variable will not decrease in value by
more than 2.326 standard deviations.
Single-Asset Case
Therefore the one-day 99% VaR for our portfolio consisting of a $10
million position in Microsoft is:
2.326 × 200,000 = $465,300

As discussed earlier, the 𝑁-day VaR is calculated as 𝑁 times the one-

day VaR. The 10-day 99% VaR for Microsoft is therefore
465,300 × 10 = $1,471,300
Example 3
Suppose the return from an investment over a specified time horizon
has a normal distribution with mean 20 and standard deviation 30. This
corresponds to a loss distribution with mean −20 and standard
deviation 30.
Solution 3
Suppose that we wish to calculate the VaR with a confidence level of
99%. This is the loss level that has a 1% chance of being exceeded. The
function NORM. INV in Excel can be used for this calculation. The first
argument is the percentile of the distribution required (0.99), the
second is the mean loss (−20), and the third is the standard deviation
(30).
NORM. INV(0.99, −20,30) gives 49.79. Thus, the 99𝑡ℎ percentile of a
normal loss distribution with a mean of −20 and a standard deviation of
30 is 49.79. This is the VaR level when the confidence level is 99%.
Example 4
Assume the result of an investment with a uniform distribution where
all outcomes between a profit of 30 and a loss of 20 are equally likely. In
this case, the VaR with a 99% confidence level is 19.5. This is because
the probability that the loss will lie between 19.5 and 20 is 0.5/50 =
1%.
(Note that we divide the range of losses we are interested in (0.5) by
the total range of losses (50) because all outcomes are equally likely.)
1% of 50 is 1% × 50 = 0.5
Example 5
Suppose that for a one-year project all outcomes between a loss of 50
million and a gain of 50 million are considered equally likely. In this case,
the loss from the project has a uniform distribution extending from –50
million to +50 million. There is a 1% chance that there will be a loss
greater than 49 million.
Solution 5
The VaR with a one-year time horizon and a 99% confidence level is
therefore 49 million.

Kita ‘confident’ 99%, kerugian (loss) tidak akan melebihi 49 juta.

Example 6
As a final example, let us suppose that the outcomes are discrete rather
than continuous. Suppose that a project with one year remaining is
performing badly and is certain to lead to a loss. Three different
scenarios are possible:
1. A loss of 2 million (probability 88%),
2. A loss of 5 million (probability 10%), and
3. A loss of 8 million (probability 2%).
Solution 6
This is summarized in table below. The final column shows the
cumulative loss probability measured from the lowest to the highest.
The first 88% of the cumulative loss distribution corresponds to a loss of
2 million. The part of the cumulative distribution between 88% and 98%
corresponds to a loss of 5 million. The part of the cumulative
distribution between 98% and 100% corresponds to a loss of 8 million.
Solution 6
Solution 6
With a 99% confidence level, the VaR is 8 million. This is because 99%
falls into the 98% to 100% cumulative probability range, and the loss
for this range is 8 million. When the confidence level is reduced to 97%,
the VaR is 5 million because 97% falls into the 88% to 98% cumulative
probability range, and the loss for this range is 5 million.
Solution 6
Note that if the confidence level is 98%, there is some ambiguity. We
could argue that we are in the 98% to 100% range so that the VaR is 8
million, or that we are in the 88% to 98% range so that the VaR is 5
million. One approach here is to set the VaR equal to the average of the
two answers, or 6.5 million.
Example 7
A one-year project has a 98% chance of leading to a gain of 2 million, a
1.5% chance of leading to a loss of 4 million, and a 0.5% chance of
leading to a loss of 10 million. The cumulative loss distribution is shown
in graphic.
Example 7
Solution 7

Loss (million) Probability (%) Cumulative Probability Range (%)

-2 98 0 to 98
4 1.5 98 to 99.5
10 0.5 99.5 to 100
Solution 7
The point on this cumulative distribution that corresponds to a
cumulative probability of 99% is 4 million. It follows that VaR with a
confidence level of 99% and a one-year time horizon is 4 million.
Example 8
Consider again the situation in Example 4. Suppose that we are
interested in calculating a VaR using a confidence level of 99.5%.
Solution 8
In this case, graphic shows that all losses between 4 million and 10
million have a probability of 99.5% of not being exceeded. Equivalently,
there is a probability of 0.5% of any specified loss level between 4
million and 10 million being exceeded. VaR is therefore not uniquely
defined. One reasonable convention in this type of situation is to set VaR
equal to the midpoint of the range of possible VaR values. This means
that, in this case, VaR would equal 7 million.
Exercise 1
A fund manager announces that the fund’s one-month 95% VaR is 6% of
the size of the portfolio being managed. You have an investment of
100,000 in the fund.
How do you interpret the portfolio manager’s announcement?
Solution 1
There is a 5% chance that you will lose 6,000 or more during a one-
month period.
Exercise 2
Suppose that the change in the value of a portfolio over a one-day time
period is normal with a mean of zero and a standard deviation of 2
million, what is
(a) The one-day 97.5% VaR ?
(b) The five-day 97.5% VaR ?
(c) The five-day 99% VaR ?
Solution 2
(a) 1 × 2 × 1.96 = 3.92 million,

(b) 5 × 2 × 1.96 = 8.77 million,

(c) 5 × 2 × 2.33 = 10.40 million.