Scribd
Upload
41 views25 pages

Value at Risk Lecture

This document discusses value at risk (VaR), a measure used to quantify the total risk of a portfolio of financial assets. It defines VaR as capturing the total risk within a set of assets and reducing it to a single number representing the maximum potential loss over a given time period at a certain confidence level. The document presents the historical simulation method for calculating VaR using past market data to estimate potential future losses. It provides an example of how historical simulation would work using data on stock index changes to estimate the potential loss of a portfolio over the next day at a 99% confidence level.

Uploaded by

jknibm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
41 views25 pages

Value at Risk Lecture

This document discusses value at risk (VaR), a measure used to quantify the total risk of a portfolio of financial assets. It defines VaR as capturing the total risk within a set of assets and reducing it to a single number representing the maximum potential loss over a given time period at a certain confidence level. The document presents the historical simulation method for calculating VaR using past market data to estimate potential future losses. It provides an example of how historical simulation would work using data on stock index changes to estimate the potential loss of a portfolio over the next day at a 99% confidence level.

Uploaded by

jknibm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 25

2.

Value at risk

Lecture slides
Introduction
• Discuss the concept of VaR in more detail

• Present two key methods for establishing


VaR
Value at risk

This is a very widely used measure designed to capture the


total risk within a set of financial assets and reduce it to a
single number…

The goal of this measure is at the end to be able to say “I


am X percent confident that this portfolio will not lose
more than Y over the next Z days”

VaR
Why is VaR important?

• The Basel Committee use it as a central requirement for


assessing a banks capital adequacy requirement

• So we are concerned with the far left of the normal


distribution curve, what is the worst case?
VaR: How bad can things get?
Expected Shortfall: If things do go bad how much can the company expect to loose?
Expected loss during a Z day period conditional on the loss being worse than VaR
The Time Horizon
Historical Simulation

• We collect past data to generate an idea about the


future.
• Collect data on market variables and the value of what it
is we are assessing, a portfolio for example.
• We generate our own distribution based on this data we
have collected.
• This distribution can then be used to estimate what the
lowest possible value the portfolio can take.
Example: Historical Simulation

• An investor in the US owns on Sept 25th 2008 a portfolio;

• What will this portfolio be worth tomorrow?

Worksheets available through the Core text


Example: Historical Simulation
Example: Historical Simulation
Example: Historical Simulation;
Scenario 1
Data on stock indices for historical simulation after exchange rate
• So, scenario 1 is the change in the values
adjustments
Day Date DIJA FTSE 100 CAC 40 Nikkei
between Day 0 and Day 1. 225
0 Aug. 7th 2006 11,219.38 11,131.84 6373.89 131.77
1 Aug 8th 2006 11,173.59 11,096.28 6378.16 134.38
Example: Historical Simulation;
Scenario 1
Example: Historical Simulation
• If weLosses
repeat this
ranked
scenarios
fromprocess for
highest to lowest all
for 500500

scenarios, we are given


Scenario number Loss the following data;
($000s)
494 477.841
339 345.435
349 282.204
329 277.041
487 253.385
227 217.974
Example: Historical Simulation
Simplicity of our approach
• We have only used four assets, a banks books would likely be
composed of tens of thousands of different positions.
• Many of these positions would also be much more
complicated instruments than simple equity indices, such as
derivatives contracts (which we will cover later on in the
module).
• This would be calculated every single day, so the process
rolls forward each time.
• Other market variables such as interest rates, different
currencies and many others could also be included.
Stress testing
• Companies have also calculated VaR based around data
when extreme market conditions have been observed.

• A 5 standard deviation shift in market variables could be


considered to be an extreme case, which under the
assumptions of the normal distribution would occur once
every 7,000 years. When in reality occurs around once or
twice every 10 years.

• We have done this in our example


Model-building approach: Single-
asset
Building on the discussion from the previous
lecture if we know a few pieces of information
we can establish the VaR by applying a few
basic principles.

All we are looking for is the tail end of the


distribution…
Example: Microsoft shares
We currently own $10,000,000 of shares in Microsoft,
and we would like to establish the 10 day 99% VaR.

Assume that the volatility is 2%, therefore $200,000

The expected change, i.e. the mean change is generally


considered to be zero, or so much smaller than the
standard deviation it hardly makes a difference to our
calculations, (I will show an example of this later).
Example: Microsoft shares
µ=0
Σ = 200,000

- - - 0 200,000 400,000 600,000


600,000 400,000 200,000
Scaling Value at Risk
• If we hold an asset for a period longer than a
day, it makes sense that the VaR would
increase. Your holding period increases
therefore there is a greater chance of
adverse price changes.
• But we couldn't simply multiply by the number
of days in the period, this wouldn't make
sense.
• The reason why starts off with a basic assumption in finance;

• The asset price returns are independent and identically


distributed with a mean of zero. Basically meaning that our
asset price returns are like flips of a coin, a more technical
description would be that they behave like particles in a fluid.

• This means that the variance of our asset price returns


remains constant and is stationary.
Summary
• VaR is our one measure for stating the
losses we could incur, very useful to distil
risk to one figure.
• Historical and single asset methodologies
• Applications
Reading on this topic, and
questions?
• Hull, J. (2012) Options, Futures, and Other
Derivatives. Chapter 21, pages 471-478.
• or 9th Edition Chapter 22, pages 494-501

You might also like