Hull RMFI4 e CH 12

Value at Risk and
Expected Shortfall
Chapter 12
Risk Management and Financial Institutions 4e, Chapter 12, Copyright John C. Hull 2015
The Question Being Asked in VaR

What loss level is such that we are X%
confident it will not be exceeded in N
business days?
VaR and Regulatory Capital

Regulators
base the capital they require

banks to keep on VaR
The market-risk capital has traditionally
been calculated from a 10-day VaR
estimated where the confidence level is
99%
Credit risk and operational risk capital are
based on a one-year 99.9% VaR
Advantages of VaR
It
captures an important aspect of risk

in a single number
It is easy to understand
It asks the simple question: How bad can
things get?
Example 12.1 (page 185)

The
gain from a portfolio during six month

is normally distributed with mean $2
million and standard deviation $10 million
The 1% point of the distribution of gains is
22.3310 or $21.3 million
The VaR for the portfolio with a six month
time horizon and a 99% confidence level
is $21.3 million.
Example 12.2 (page 186)

All
outcomes between a loss of $50 million

and a gain of $50 million are equally likely
for a one-year project
The VaR for a one-year time horizon and a
99% confidence level is $49 million
Examples 12.3 and 12.4 (page 186)

A
one-year project has a 98% chance of

leading to a gain of $2 million, a 1.5%
chance of a loss of $4 million, and a 0.5%
chance of a loss of $10 million
The VaR with a 99% confidence level is $4
million
What if the confidence level is 99.9%?
What if it is 99.5%?
Cumulative Loss Distribution for

Examples 12.3 and 12.4 (Figure 12.3, page
186)
VaR vs. Expected Shortfall
VaR is the loss level that will not be exceeded with a

specified probability
Expected shortfall (ES) is the expected loss given that
the loss is greater than the VaR level (also called C-VaR
and Tail Loss)
Regulators have indicated that they plan to move from
using VaR to using ES for determining market risk capital
Two portfolios with the same VaR can have very different
expected shortfalls
Distributions with the Same VaR but

Different Expected Shortfalls
VaR
VaR
10
Coherent Risk Measures (page 188)
Define a coherent risk measure as the amount of

cash that has to be added to a portfolio to make its
risk acceptable
Properties of coherent risk measure
If one portfolio always produces a worse outcome than

another its risk measure should be greater
If we add an amount of cash K to a portfolio its risk measure
should go down by K
Changing the size of a portfolio by should result in the risk
measure being multiplied by
The risk measures for two portfolios after they have been
merged should be no greater than the sum of their risk
measures before they were merged
11
VaR vs Expected Shortfall

VaR
satisfies the first three conditions but

not the fourth one
ES satisfies all four conditions.
12
Example 12.5 and 12.7
Each of two independent projects has a probability 0.98

of a loss of $1 million and 0.02 probability of a loss of
$10 million
What is the 97.5% VaR for each project?
What is the 97.5% expected shortfall for each project?
What is the 97.5% VaR for the portfolio?
What is the 97.5% expected shortfall for the portfolio?
13
Examples 12.6 and 12.8
A bank has two $10 million one-year loans. Possible outcomes are
as follows
Outcome
Probability
Neither Loan Defaults
97.5%
Loan 1 defaults, loan 2 does not default
1.25%
Loan 2 defaults, loan 1 does not default
1.25%
Both loans default
0.00%
If a default occurs, losses between 0% and 100% are equally likely.

If a loan does not default, a profit of 0.2 million is made.
What is the 99% VaR and expected shortfall of each project
What is the 99% VaR and expected shortfall for the portfolio
14
Spectral Risk Measures

1.
2.
3.
4.
A spectral risk measure assigns weights to

quantiles of the loss distribution
VaR assigns all weight to Xth percentile of the
loss distribution
Expected shortfall assigns equal weight to all
percentiles greater than the Xth percentile
For a coherent risk measure weights must be a
non-decreasing function of the percentiles
15
Normal Distribution Assumption
When losses (gains) are normally distributed

with mean and standard deviation
VaR N 1 ( X )
ES
Y 2 2
e
2 (1 X )
16
Changing the Time Horizon

If
losses in successive days are

independent, normally distributed, and
have a mean of zero
T - day VaR 1 - day VaR T
T - day ES 1 - day ES T
17
Extension
there is autocorrelation between the
losses (gains) on successive days, we
replace T by
If
T 2(T 1) 2(T 2) 2 2(T 3) 3 2 T 1
in these equations
18
Ratio of T-day VaR to 1-day VaR (Table 12.1,

page 193)
T=1
T=2
T=5
T=10
T=50
T=250
1.0
1.41
2.24
3.16
7.07
15.81
=0.05 1.0
1.45
2.33
3.31
7.43
16.62
=0.1 1.0
1.48
2.42
3.46
7.80
17.47
=0.2 1.0
1.55
2.62
3.79
8.62
19.35
=0
19
Choice of VaR Parameters

Time horizon should depend on how quickly
portfolio can be unwound. Regulators are
planning to move toward a system where ES is
used and the time horizon depends on liquidity.
(See Fundamental Review of the Trading Book)
Confidence level depends on objectives.
Regulators use 99% for market risk and 99.9%
for credit/operational risk.
A bank wanting to maintain a AA credit rating
might use confidence levels as high as 99.97%
for internal calculations.
20
Aggregating VaRs
An approximate approach that seems to works
well is
VaR total
VaR VaR
i
ij
where VaRi is the VaR for the ith segment,

VaRtotal is the total VaR, and ij is the coefficient
of correlation between losses from the ith and
jth segments
21
VaR Measures for a Portfolio where an

amount xi is invested in the ith component
of the portfolio (page 195-196)
Marginal
VaR:
VaR
xi
Incremental
VaR: Incremental effect of

the ith component on VaR
Component
VaR:
VaR
xi
xi
22
Properties of Component VaR

The
component VaR is approximately the

same as the incremental VaR
The total VaR is the sum of the component
VaRs (Eulers theorem)
The component VaR therefore provides a
sensible way of allocating VaR to different
activities
23
Back-testing (page 197-200)
Back-testing a VaR calculation methodology involves

looking at how often exceptions (loss > VaR) occur
Alternatives: a) compare VaR with actual change in
portfolio value and b) compare VaR with change in
portfolio value assuming no change in portfolio
composition
Suppose that the theoretical probability of an exception
is p (=1X). The probability of m or more exceptions in n
days is
n
n!
k
nk
p
(
1
p
)
k m k!( n k )!
24
Bunching
Bunching
occurs when exceptions are not

evenly spread throughout the back testing
period
Statistical tests for bunching have been
developed by Christoffersen (See page 200)
25

Hull RMFI4 e CH 12

Uploaded by

Copyright:

Available Formats

Hull RMFI4 e CH 12

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hull RMFI4 e CH 12

Uploaded by

Copyright:

Available Formats

What is VaR and how is it used by regulators?

What is VaR and how is it used by regulators?

What are some advantages and disadvantages of using VaR?

What are some advantages and disadvantages of using VaR?

Value at Risk and

The Question Being Asked in VaR

VaR and Regulatory Capital

base the capital they require

captures an important aspect of risk

Example 12.1 (page 185)

gain from a portfolio during six month

Example 12.2 (page 186)

outcomes between a loss of $50 million

Examples 12.3 and 12.4 (page 186)

one-year project has a 98% chance of

Cumulative Loss Distribution for

VaR vs. Expected Shortfall

VaR is the loss level that will not be exceeded with a

Distributions with the Same VaR but

Coherent Risk Measures (page 188)

Define a coherent risk measure as the amount of

If one portfolio always produces a worse outcome than

VaR vs Expected Shortfall

satisfies the first three conditions but

Example 12.5 and 12.7

Each of two independent projects has a probability 0.98

Examples 12.6 and 12.8

Neither Loan Defaults

Loan 1 defaults, loan 2 does not default

Loan 2 defaults, loan 1 does not default

Both loans default

If a default occurs, losses between 0% and 100% are equally likely.

Spectral Risk Measures

A spectral risk measure assigns weights to

Normal Distribution Assumption

When losses (gains) are normally distributed

Changing the Time Horizon

losses in successive days are

T 2(T 1) 2(T 2) 2 2(T 3) 3 2 T 1

Ratio of T-day VaR to 1-day VaR (Table 12.1,

Choice of VaR Parameters

where VaRi is the VaR for the ith segment,

VaR Measures for a Portfolio where an

VaR: Incremental effect of

Properties of Component VaR

component VaR is approximately the

Back-testing (page 197-200)

Back-testing a VaR calculation methodology involves

occurs when exceptions are not

You might also like