correlation and dependence. It explains that correlation is designed to measure one partic-
ular type of dependence, namely linear dependence. The chapter explains that correlation
can be monitored in the same way as volatility. Either the exponentially weighted moving
average or GARCH (1,1) procedure can be used. It also explains the positive-semidefinite
condition for a consistent variance-covariance matrix.
Copulas are explained in this chapter because the Gaussian copula model of defaults is
necessary for an understanding of Basel II in Chapter 15. As preparation for the material
on copulas, the chapter discusses the properties of multivariate normal distributions and
explains factor models. The latter provide a convenient way of defining the correlation
matrix for a large number of variables. The capital asset pricing model which is likely to
be somewhat familiar to a lot of students provides an example of a one-factor model.
The notation used by mathematicians for copulas is too abstract for many students
of risk management. As will be evident from Chapter 11 and the slides that go with it, I
prefer to use a minimal amount of algebra. I go through the copula material fairly slowly.
(I find the copula material, once understood, is considered to be so straightforward that
students cannot understand why they had any difficulty with it in the first place!)
I start by asking students to suppose they know the marginal (unconditional) probabil-
ity distributions of two variables and want to define a correlation structure. If the variables
are normally distributed it is natural to assume that they are bivariate normal (although
this is by no means the only possibility). In other cases we can map the distributions on
a “percentile-to-percentile” basis to standard normal distributions and assume that these
are bivariate normal. Thus the correlation structure between the original distributions is
defined indirectly using other easier-to-deal-with distributions.
This procedure is illustrated in the text for two variables that have triangular dis-
tributions. (I choose triangular distributions because it is easy to calculate cumulative
probabilities for them.) The marginal distributions of the variables are shown in Figure
11.2. The transformation of the first variable to a standard normal distribution is shown in
Table 11.1 and that for the second variable is shown in Table 11.2. The joint distribution
of the variables when the correlation between the normal distributions is 0.5 is shown in
Table 11.3. The overall methodology is shown diagrammatically in Figure 11.3.
Figures 11.2 and 11.3 together with Tables 11.1, 11.2, and 11.3 provide an example of
the Gaussian copula model. Other copula models can be understood similarly. For example
we can transform two variables to Student t-distributions on a percentile-to-percentile basis
and assume the transformed variables have a bivariate Student t-distribution. As shown
in Figures 11.4 and 11.5 the Student t-copula has more tail dependence than the Gaussian
copula.
When many variables are involved and the Gaussian copula model is used, each vari-
able can be transformed into a normal distribution on a percentile-to-percentile basis. A
factor model can then be used to define the correlations between the normal distributions.
Gordy’s result has been moved to Chapter 15 in the fourth edition. The proof of
Vasicek’s result is now in a separate section from a statement of the result which I think
makes material flow better.
Problems 11.16 to 11.21 all make good hand-in assignment questions.
Chapter 12: Value at Risk and Expected Shortfall
7
This chapter is similar to Chapter 9 of the third edition, but in keeping with its growing
importance there is more discussion of expected shortfall (ES). The chapter requires 2 to
3 hours of classroom time. It starts by explaining value at risk (VaR). It compares VaR
to ES and shows that the latter has better theoretical properties. (To use the technical
term, it is more “coherent.”) It discusses the choice of parameters for VaR and ES, and
the impact of autocorrelation on estimates.
One thing I found after I wrote the first edition of the book is that many students have
difficulty calculating VaR and expected shortfall when the probability distribution of losses
is discrete. Consequently, I included several extra examples illustrating the calculations.
I find it useful to go through most of these examples in class. Many financial institutions
like the properties of expected shortfall and use it internally as one of their risk measures.
The Fundamental Review of the Trading Book (see Chapter 17) indicates that ES will
soon become the regulator’s measure of choice for market risk even though it is considered
more difficult to backtest than VaR.
Sections 12.7 to 12.9 explain marginal VaR, incremental VaR, and component VaR.
They explain the role of Euler’s theorem in the allocation of VaR and give a way of
aggregating VaRs. The material on back-testing in Section 12.10 gives some standard
tests of statistical significance and includes a relatively powerful two-tailed test proposed
by Kupiec.
Any of Further Questions can be used as hand-in assignments. I usually use 12.13.
Chapter 13: Historical Simulation and Extreme Value Theory
This chapter is similar to Chapter 14 of the third edition. It has more emphasis on
expected shortfall. The sequence of chapters has been improved for the fourth edition with
the calculation of VaR and ES coming immediately after the introduction of the measures
in Chapter 12. This chapter requires about 1.5 hours of classroom time. The advantage
of the historical simulation approach is that it requires no assumptions about probability
distributions and correlations. It assumes that percentage changes in all market variables
over the next day are a random sample from the last Ndays (or in the case of stressed
VaR a random sample from consecutive days during a period of particular stress for the
portfolio). Results are presented for a simple portfolio involving investments in four indices.
Section 13.1 explains the basic historical simulation approach while Section 13.2 ex-
plains how to calculate a standard error for VaR. (The standard error is quite large and
would be even larger if the assumption that the joint distribution of daily changes in market
variables is stationary through time could be relaxed.) Section 13.3 describes some exten-
sions of the historical simulation approach. These involve alternative weighting schemes,
procedures involving the volatility updating procedures of Chapter 10, and the use of
the bootstrap method to determine a confidence interval for VaR. (New material in the
fourth edition includes a simpler application of volatility updating.) Section 15.4 covers
the delta/gamma approximation. Section 13.5 and 13.6 present material on extreme value
theory which extends the material on the power law in Chapter 10. It provides a scientific
way of “smoothing the tails” of an empirically observed distribution.
Excel worksheets that go with the four-index example that is covered in the chapter
can be downloaded from my web site.
8