Modelling Extremes in Insurance and Finance:
Practical Necessity versus Methodological
Challenge
Paul Embrechts
Department of Mathematics
ETH Zurich
Switzerland
www.math.ethz.ch/~embrechts
28th ICA Paris, May 29, 2006
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 1 / 51
What is the importance of the date
February 1, 1953
for Quantitative Risk Management in general and the modelling of
extremes more in particular?
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 2 / 51
“Springtij en orkaan veroorzaken nationale ramp.
Nederland in grote watersnood.”
(De Yssel - en Lekstreek, 6/2/53)
January 31 - February 1, 19531 (February flood):
• 1 836 people killed
• 72 000 people evacuated
• 49 000 houses and farms flooded
• 201 000 cattle drowned
• 500 km of sea dykes destroyed, 400 breaches
• 200 000 ha of land flooded
1
Antwerp (Schoten), February 3, 1953
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 3 / 51
En in alle gewesten
Wordt de stem van het water
Met zijn eeuwige rampen
Gevreesd en gehoord
(Marsman, 1938)
Throughout the centuries, the Dutch coast has been suffering
numerous disastrous floods:
1421, 1570, 1775, 1825, 1916, 1925, 1953, 20022
1421: The St. Elisabeth flood
1570: The All Saints flood
2
Not complete
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 4 / 51
The Delta Project
• coastal defence against flooding
• required dyke height at location l : hδ (l )
• safety margin: MYSH(l ) = Maximum Yearly Sea Height at
location l :
P(MYSH(l ) > hδ (l )) “small”
where “small” means:
I 10−4 at l = Randstad
I 4−1 × 10−3 at l = Delta area and the North
• similar safety requirements for rivers but with safety values in
the range of 10−3 − 10−2
• solution for l = Randstad: hδ (l ) = NAP + 5.14 m
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 5 / 51
Hence, we can reformulate the problem as:
• “small” = 1 − α ∈ (0, 1), α ≈ 1
➟ “small”: Extreme Value Theory (EVT)
• l = fixed
➟ different l : copulae/as
• MYSH(l ) = X , a rv with df F
➟ co-variables: EVT, copulae, non-stationarity
• hδ (l ) = qα
qα = F −1 (α) (α − quantile)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 6 / 51
The February flood happened as
a coincidence of several factors
• very low pressure over
Scotland
• very strong North-Westerly
storm
• “springtij” (extremely high
tide)
All contributing to “the perfect storm scenario”
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 7 / 51
Conclusions from the dyke example so far:
• extreme events happen
• use historical data and other information to model the
underlying risk process
• agree on a quantitative measure of risk
• estimate as well as possible
• live by the conclusions
• revisit analysis regularly and critically
• do not forget about qualitative issues
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 8 / 51
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 9 / 51
This was an example from the realm of
environmental extremes
but what about
extremes in insurance and finance?
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 10 / 51
Some quotes
Press Release
January 20, 2006
The Committee of European Banking Supervisors on the
validation and assessment of credit and operational risk
approaches
“If an Extreme Value Theory approach is used, resorting
to the stability property of the model (the so-called
‘Peaks over Threshold’ stability property) that makes it
possible to compute the highest percentiles of the
frequency and severity distributions from figures
estimated at a lower level (usually at the threshold
level).”
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 11 / 51
“The term correlation as used in the CRD, should be
interpreted broadly to mean any form of dependency
(e.g. linear or non-linear)”
“Dependencies between tail events should be studied
with great care.”
“The credit institution must validate its correlation
assumptions using appropriate quantitative and
qualitative techniques.”
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 12 / 51
GRIT Report
March 27, 2006
Institute of Actuaries and Faculty of Actuaries:
A Change Agenda for Reserving
“ Identifying where our Reserving Methods need to be
Enhanced:
More sophisticated mathematical and statistical
methodology need not be a priority for actuaries at this
stage. Rather, the focus for enhancement and research
should be in the following areas:
- ...
- Whether extreme value theory has a role to play in
reserving
- ...”
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 13 / 51
Report of Solvency Working Party Prepared for IAA Insurance
Regulation Committee:
- Strong emphasis on the treatment of extreme event risks
- Potential use of EVT (even including an analysis of the
Danish Fire Insurance data, p. 61-62!)
- The report emphasizes also the problem of modelling
dependence (p. 33-35)
- In several subsequent reports, the notion of copula enters
explicitly
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 14 / 51
“A natural consequence of the existence of a lender of
last resort is that there will be some sort of allocation of
burden of risk of extreme outcomes. Thus, central banks
are led to provide what essentially amounts to
catastrophic insurance coverage ... From the point of
view of the risk manager, inappropriate use of the
normal distribution can lead to an understatement
of risk, which must be balanced against the significant
advantage of simplification. From the central banks
corner, the consequences are even more serious because
we often need to concentrate on the left tail of the
distribution in formulating lender-of-last-resort policies.
Improving the characterization of the distribution of
extreme values is of paramount importance.”
(Alan Greenspan, Joint Central Bank Research Conference, 1995)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 15 / 51
“Extreme, synchronized rises and falls in financial
markets occur infrequently but they do occur. The
problem with the models is that they did not assign a
high enough chance of occurrence to the scenario in
which many things go wrong at the same time - the
“perfect storm” scenario”
(Business Week, September 1998)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 16 / 51
“Regulators have criticised LTCM and banks for not
“stress-testing” risk models against extreme market
movements... The markets have been through the
financial equivalent of several Hurricane Andrews hitting
Florida all at once. Is the appropriate response to accept
that it was mere bad luck to run into such a rare event -
or to get new forecasting models that assume more
storms in the future?”
(The Economist, October 1998, after the LTCM rescue)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 17 / 51
“... The trading floor is quiet. But this masks their
attempt at picking up the pieces with a new fund, JWM
Partners. Now, Mr. Meriwether is preaching new gospel:
World financial markets are bound to hit extreme
turbulences again... Mr. Meriwether’s crew, once
bitten, also is betting on more liquid securities: “With
globalisation increasing, you’ll see more crises,” he says.
“Our whole focus is on the extremes now - what’s the
worst that can happen to you in any situation -
because we never want to go through that again” ”
(John Meriwether, The Wall Street Journal, 21/8/2000)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 18 / 51
Hence:
Practical Necessity
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 19 / 51
EKM QRM
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 20 / 51
EKM reviews the one-dimensional theory of
extremes:
Mn = max(X1 , . . . , Xn )
and yields models like:
• generalized extreme value distributions
• generalized Pareto distributions
in order to estimate quantities like
• long return periods
• high quantiles (like Value-at-Risk)
• risk measures beyond VaR
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 21 / 51
An example: the Danish fire insurance data
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 22 / 51
The EVT-POT analysis
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 23 / 51
Hence:
Methodological Challenge
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 24 / 51
Statements on the use of EVT methodology
“There is always going to be
an element of doubt, as one is
extrapolating into areas one
doesn’t know about. But
what EVT is doing is making
the best use of whatever data
you have about extreme
phenomena.”
(Richard L. Smith)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 25 / 51
Statements on the use of EVT methodology
“The key message is that
EVT cannot do magic – but it
can do a whole lot better than
empirical curve fitting and
guesswork. My answer to the
skeptics is that if people aren’t
given well-founded methods
like EVT, they’ll use dubious
ones instead.”
(Jonathan E. Tawn)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 26 / 51
Many applications:
• environmental
• records
• catastrophe bonds
• energy pricing
• mining/geology
• reinsurance...
But in particular in
Quantitative Risk Management
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 27 / 51
Extremes manifest themselves in many ways in
insurance and finance:
• extreme losses (size and frequency)
• extreme price movements
• extreme risk measures
• extreme correlations
• diversification breakdown
• spillover
• systemic risk ...
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 28 / 51
Two examples with implicit extremal problems
Example 1:
“Life as a risk manager is very dangerous if you only wear
a pair of correlation-marginal glasses.”
Example 2:
“Please be aware that linear correlation and marginal
distributions speak to each other.”
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 29 / 51
Example 1:
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 30 / 51
A risk manager’s life can get worse:
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 31 / 51
Example 2:
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 32 / 51
Learn about copulas
• X1 , . . . , Xd one-period risks,
P(Xi ≤ x) = Fi (x), i = 1, . . . , d
assumed continuous
• then there exists a unique function C (the copula) so that
F (x) = P(X1 ≤ x1 , . . . , Xd ≤ xd ) = C (F1 (x1 ), . . . , Fd (xd ))
• Schematically:
F ⇔ (C ; F1 , . . . , Fd )
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 33 / 51
Cookbook
F ⇒ (C ; F1 , . . . , Fd )
Stress Testing
F ⇐ (C ; F1 , . . . , Fd )
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 34 / 51
• Some history
• Some highlights
• Some caution
• Some future
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 35 / 51
Some history
Pre 1950 Fréchet, Hoeffding, Dall’Aglio, Féron
1959 Sklar, Schweizer
After 1970 Schweizer, Wolff, Nelsen, Scarsini, Joe, Rüschendorf,
Genest, Rivest, ...
Textbooks on the methodology
• Joe (1997)
• Nelsen (1999, 2006)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 36 / 51
Applications to insurance, finance and QRM
• Luciano (early 90s)
• Frees and Valdez (1997)
• Embrechts, McNeil and Straumann (1998, RiskLab report)
• Li (1998,...)
• Klugmann and Parsa (1999)
• McNeil, Frey and Embrechts (2005)
• search for copula, risk management gave
75 000 hits on May 18, 2006 (71 000 for extreme value
theory, risk management)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 37 / 51
Some highlights
• Limitations of linear correlation
• Alternative measures of dependence
• Models for credit risk (CDOs, ...)
• Stress testing
My personal view:
pedagogic, pedagogic, stress testing
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 38 / 51
Some criticism
“My main concern is that this
very simple concept might be
something like the emperor’s
new clothes because it
promises to solve all problems
of stochastic dependence but
it falls short in achieving the
goal.”
(Thomas Mikosch)
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 39 / 51
Some future
• Copulas have outgrown their infancy
• Serious applications emerge
• Need for global modelling in high dimensions
• Need for more realistic dynamic models
• Increasing awareness that copulas are just an extra tool, not a
panacea for multivariate modelling
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 40 / 51
Current work in QRM at RiskLab, ETH Zurich
• Operational Risk
• Risk Aggregation
• Multivariate Extremes
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 41 / 51
Operational Risk
Definition (Basel II):
The risk of losses resulting from inadequate or failed internal
processes, people and systems, or external events.
This definition includes legal risk, but excludes strategic and
reputational risk.
Example:
Barings Bank, February 26, 1995
Similar discussion:
Solvency 2 for insurance
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 42 / 51
The LDA approach to Pillar I
• Definition: VaR = VaR(1 year, 99.9%), hence EVT
• Estimation: VaR1 , . . . , VaRd
• Aggregation: VaR+ = VaR1 + · · · + VaRd
• Diversification: VaROP < VaR+ , hence copulas
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 43 / 51
Risk aggregation
Marginal risk measures:
VaRMR , VaRCR , VaROR , VaRUR , ..
Global risk measure: How to aggregate and scale?
Example: How to combine a 99% - 10 day VaRMR with a 99.9%
-1 year VaROR ?
Economic capital: VaR(99.97%, 1 year)
Not at all easy!
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 44 / 51
After all these more applied issues, it is time for some
mathematics:
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 45 / 51
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 46 / 51
A one-picture summary
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 47 / 51
Research issues
• Rare event simulation
• Curse of dimensionality
• Dynamic models
• Discrete models
• Extremes versus Outliers
• Robustness
• Judgment, ...
• Data – Data – Data
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 48 / 51
Conclusions
• Let us not forget: risk also concerns the upside
• Concerning the downside
- extremes matter (practice)
- extremal-type questions are asked (business, clients,
regulators)
- extreme value theory offers a canonical tool which has to be
used with great care
- extremal behavior not only concerns severity, but also
dependence
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 49 / 51
• These type of questions and resulting methodology are key to
QRM and ERM and consequently should be standard
knowledge to any actuary, hence
training
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 50 / 51
c P. Embrechts (ETH Zurich)
Modelling Extremes in Insurance and Finance 28th ICA Paris 51 / 51
