Market Risk

Measurement
Lecture 0:
Course Organization

Romain DEGUEST

Market Risk Measurement

MSc in Financial Markets

Romain DEGUEST
(EDHEC-Risk Institute)

romain.deguest@edhec.edu

Lecture 0:
Course Organization

1/12
 Market Risk
Introduction Measurement
Lecture 0:
Course Organization

Romain DEGUEST

The subject of this course is risk measurement and risk management.

The main objectives of the course is to answer the following questions:

What is risk and what are the different sources of risk?
How can we measure the risk of a general portfolio?
How can we manage the different types of risk?

The main challenge is to present the ideas in a consistent way.

Although there are fundamental theories we will use, there is no unified

theory of risk measurement and management as such.

2/12
 Market Risk
General Structure Measurement
Lecture 0:
Course Organization

Romain DEGUEST
The course consists of 10 lectures covering a wide range of topics. The
first block of 15 hours is divided as follows:

Lecture 1: Introduction to risks, market risk definition, lessons

from the past, regulation, risk management, P&L distribution,
quantiles of distribution, Value-at-Risk (VaR ) definition;
Lecture 2: Volatility estimation, implied volatility, review of
econometric concepts, moving average model, exponentially
weighted moving average model, GARCH model, (case study with
a comparison of the models);
Lecture 3: VaR computation techniques, historical VaR , hybrid
historical VaR , parametric VaR , Monte Carlo VaR computation,
delta-normal approach;
Lecture 4: Marginal VaR , Incremental VaR , Component VaR ,
Back-testing of VaR using Kupieck’s and Christoffersen’s tests;
Lecture 5: Case study: VaR computation with Matlab (in class
graded project).

3/12
 Market Risk
General Structure Measurement
Lecture 0:
Course Organization
The second block of 15 hours consists of: Romain DEGUEST

Lecture 6: Expected Shortfall (ES ), coherent risk measures,

convex risk measures, the coherence axioms and their application,
spectral risk measure, beyond Gaussian models, Cornish-Fisher
approximation of VaR , Extreme Value Theory, robustness of risk
measurement procedures;
Lecture 7: Static portfolio theory, mean-variance portfolio, efficient
frontier, replacing variance criteria by ES , dynamic asset
allocation, risk control in dynamic allocations, constant proportion
portfolio insurance (CPPI) strategies, Solvency II regulation;
Lecture 8: Operational risk, liquidity risk, model risk:
understanding and quantifying these alternate risks (case study
with model risk in option pricing);
Lecture 9: Dependence modeling, multivariate distribution,
dimension reduction, Principal Components Analysis (PCA)
method;
Lecture 10: Case study: ES computation and Monte Carlo risk
measure computation with matlab (in class graded project).

4/12
 Market Risk
Schedule Measurement
Lecture 0:
Course Organization

Romain DEGUEST

We will meet 10 times:

Tuesday January 13th (lecture 1: 9:30-12:15),

Wednesday January 14th (lecture 2: 9:30-12:15),
Thursday January 15th (lecture 3: 9:30-12:15),
Tuesday January 20th (lecture 4: 9:30-12:15),
Monday January 26th (lecture 5: 9:00-13:00),
Wednesday February 4th (lecture 6: 9:30-12:15),
Thursday February 5th (lecture 7: 9:30-12:15),
Tuesday February 10th (lecture 8: 9:30-12:15),
Wednesday February 11th (lecture 9: 14:00-16:45),
Tuesday February 17th (lecture 10: 9:00-13:00),

5/12
 Market Risk
Organization Measurement
Lecture 0:
Course Organization

Romain DEGUEST

In this course, several case studies, computer programming practices

will be provided so that you become familiar with the concepts and
procedures that are applied in the financial industry.

The final grade will be based on the three following evaluations:

15% Case study 1 (in class with matlab)
15% Case study 2 (in class with matlab)
70% Final Exam (in class)

6/12
 Market Risk
References Measurement
Lecture 0:
Course Organization

Romain DEGUEST

These lecture notes are based on the material covered by Pr. Stoyanov
in his course given to Executive MSc students, and also follow several
chapters from:

Jorion, P. (2007) Value-at-Risk, McGraw–Hill.

McNeil, A, R. Frey, and P. Embrechts (2005) Quantitative risk
management: Concepts, techniques, and tools, Princeton
University Press.
Dowd, K. (2005) Measuring market risk, Wiley.

7/12
 Market Risk
Prerequisites: Basic Rules for Integration Measurement
Lecture 0:
Course Organization

Romain DEGUEST

The integral of a function f is defined informally to be the net

signed area of the region in the xy −plane bounded by the graph of
f , the x−axis, and the vertical lines x = a and x = b:

f is defined as a function of the well-behaved variable x;

dx represents an infinitesimal increment of the variable x;

8/12
 Market Risk
Prerequisites: Basic Rules for Differentiation Measurement
Lecture 0:
Course Organization
The derivative of a function at a chosen input value describes the
Romain DEGUEST
best linear approximation of the function near that input value.
For a real-valued function of a single variable x, the derivative at a
point equals the slope of the tangent line to the graph of the
function at that point.

Rx
Recall that f (x) − f (0) = 0 f 0 (s)ds.
For a function f of two variables x and y , we call partial derivatives
the two quantities:
∂ ∂
f (x, y ) f (x, y )
∂x ∂y
Sometimes, we write these partial derivatives: fx (x, y ) and fy (x, y ).
9/12
 Market Risk
Prerequisites: Knowledge of Probability Measurement
Lecture 0:
Course Organization
Probability density function (PDF) f : it describes the relative
Romain DEGUEST
likelihood for a continuous random
Rb variable to occur at a given
point, i.e. P(a ≤ X ≤ b) = a f (s)ds.
Intuitively, f (s)ds represents the probability that X belongs to an
infinitesimal region of size ds around s, i.e.
f (s)ds = P(s ≤ X ≤ s + ds).
The Gaussian PDF with mean µ, variance σ 2 is simply given by the
following function:
1 −(x−µ)2
f (x) = √ e 2σ2
2πσ 2

10/12
 Market Risk
Prerequisites: Knowledge of Probability Measurement
Lecture 0:
Cumulative distribution function (CDF) F : it describes the Course Organization

probability that a real-valued random variable X with a given PDF Romain DEGUEST

f will be found at a value

R x less than or equal to x, i.e.
F (x) = P(X ≤ x) = −∞ f (s)ds.
By definition, the values taken by any CDF are always between 0
and 1.
The CDF of a standard Gaussian random variable is often denoted
N, i.e. N(x) = P(X ≤ x), where X ∼ N (0, 1).

11/12
 Market Risk
Prerequisites: Knowledge of Probability Measurement
Lecture 0:
Course Organization

Romain DEGUEST
R∞
E[g (X )] = −∞
g (x)f (x)dx.
R∞
In particular, the mean is given by E[X ] = −∞
xf (x)dx.
The variance is defined as:

Var(X ) = E[(X − E[X ])2 ] = E[X 2 ] − E[X ]2 .

X −µ
For a Gaussian variable X ∼ N (µ, σ 2 ), we have σ
∼ N (0, 1).
For two independent Gaussian variables X ∼ N (µX , σX2 ), and
Y ∼ N (µY , σY2 ) we have,

aX + bY ∼ N (aµX + bµY , a2 σX2 + b 2 σY2 )

For two jointly

 Gaussian
  variables
σX2
  
X µX ρσX σY
∼N , , we have
Y µY ρσX σY σY2

aX + bY ∼ N (aµX + bµY , a2 σX2 + b 2 σY2 + 2ρaσX bσY )

12/12