Professional Documents
Culture Documents
Solvency II
Chapter 11
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
History of Bank Regulation
l Pre-1988
l 1988: BIS Accord (Basel I)
l 1996: Amendment to BIS Accord
l 1999: Basel II first proposed
l 2003: the New Basel Capital Accord
l 2004: tentative Agreement
l 2006: Basel II
l 2009: Basel III proposal
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
The Model used by Regulators
(Figure 11.1, page 235)
Expected X% Worst
Loss Case Loss
Required
Capital
0 1 2 3 4
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Pre-1988
l Banks were regulated using balance sheet measures
such as the ratio of capital to assets
l Definitions and required ratios varied from country to
country
l Enforcement of regulations varied from country to
country
l Bank leverage increased in 1980s
l Off-balance sheet derivatives trading increased
l LDC debt was a major problem
l Basel Committee on Bank Supervision set up
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
1988: BIS Accord (page 223)
l The assets:capital ratio must be less than
20. Assets includes off-balance sheet
items that are direct credit substitutes such
as letters of credit and guarantees
l Cooke Ratio: Capital must be 8% of risk
weighted amount. At least 50% of capital
must be Tier 1.
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Example of Simple Bank Balance
Sheet: End 2012 (Table 2.2, page 24)
Assets Liabilities
Cash 5 Deposits 90
Marketable Securities 10 Subord L.T. Debt 5
Loans 80 Equity Capital 5
Fixed Assets 5
Total 100 Total 100
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Types of Capital (page 225-226)
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Risk-Weighted Capital
l A risk weight is applied to each on-balance- sheet
asset according to its risk (e.g. 0% to cash and govt
bonds; 20% to claims on OECD banks; 50% to
residential mortgages; 100% to corporate loans,
corporate bonds, etc.)
l For each off-balance-sheet item we first calculate a
credit equivalent amount and then apply a risk weight
l Risk weighted amount (RWA) consists of
l sum of risk weight times asset amount for on-balance sheet
items
l Sum of risk weight times credit equivalent amount for off-
balance sheet items
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Credit Equivalent Amount
l The credit equivalent amount is calculated
as the current replacement cost (if
positive) plus an add on factor
l The add on amount varies from instrument
to instrument (e.g. 0.5% for a 1-5 year
swap; 5.0% for a 1-5 year foreign currency
swap)
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Add-on Factors (% of Principal)
Table 11.2, page 225
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
The Math
N M
RWA = å wi Li + å w C j *
j
i =1 j =1
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
G-30 Policy Recommendations
(page 226-227)
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Netting (page 227-228)
l Netting refers to a clause in derivatives
contracts that states that if a company
defaults on one contract it must default on
all contracts
l In 1995 the 1988 accord was modified to
allow banks to reduce their credit
equivalent totals when bilateral netting
agreements were in place
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Netting Calculations
l Without netting exposure is
N
å max(V ,0)
j =1
j
l NRR =
Exposure with Netting
Exposure without Netting
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Netting Calculations continued
l Credit equivalent amount modified from
N
å[max(V ,0) + a L ]
j =1
j j j
l To
N N
max(åV j ,0) + å a j L j (0.4 + 0.6 ´ NRR )
j =1 j =1
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
1996 Amendment (page 229-231)
l Implemented in 1998
l Requires banks to measure and hold
capital for market risk for all instruments in
the trading book including those off
balance sheet (This is in addition to the
BIS Accord credit risk capital)
l It also outlined a Standardized approach,
internal model based approach (IMA)
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
1996 Amendment (page 229-231)
l The VaR Calculations reflects movements
in interest rates, exchange rates, stock
index and commodity prices but not
company specific risks. Movement in stock
prices or credit spread is captured by
Specific Risk Charges (SRC).
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
The Market Risk Capital
l The capital requirement is k ´ VaR + SRC
l Max(VaRt-1, mc X VaRavg )+ SRC
l Where k is a multiplicative factor chosen
by regulators (at least 3), VaR is the 99%
10-day value at risk, and SRC is the
specific risk charge for idiosyncratic risk
related to specific companies
l If only one-day VaR is measured then use mc X
3.16
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Basel II
l Implemented in 2007
l Three pillars
l New minimum capital requirements for credit
and operational risk
l Supervisory review: more thorough and
uniform
l Market discipline: more disclosure
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
New Capital Requirements
l Risk weights based on either external
credit rating (standardized approach) or a
bank’s own internal credit ratings (IRB
approach)
l Recognition of credit risk mitigants
l Separate capital charge for operational
risk
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
USA vs European Implementation
l In US Basel II applies only to large
international banks
l Small regional banks required to
implement “Basel 1A’’ (similar to Basel I),
rather than Basel II
l European Union requires Basel II to be
implemented by securities companies as
well as all banks
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
New Capital Requirements
Standardized Approach, Table 11.3, page 233
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
New Capital Requirements
IRB Approach for corporate, banks and sovereign
exposures
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Key Model (Gaussian Copula)
é N -1 ( PD) + r ´ N -1 (0.999) ù
WCDR = N ê ú
êë 1- r úû
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Numerical Results for WCDR
Table 11.4, page 236
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Dependence of r on PD
l For corporate, sovereign and bank
exposure
1 - e -50´PD é 1 - e -50´PD ù -50´ PD
r = 0.12 ´ -50
+ 0.24 ´ ê1 - -50 ú = 0.12[1 + e ]
1- e ë 1- e û
PD 0.1% 0.5% 1.0% 1.5% 2.0%
WCDR 3.4% 9.8% 14.0% 16.9% 19.0%
1 + ( M - 2.5) ´ b
Capital = EAD ´ LGD ´ (WCDR - PD ) ´
1 - 1. 5 ´ b
where M is the effective maturity and
b = [0.11852 - 0.05478 ´ ln( PD )]2
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Retail Exposures
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Credit Risk Mitigants
l Credit risk mitigants (CRMs) include
collateral, guarantees, netting, the use of
credit derivatives, etc
l The benefits of CRMs increase as a bank
moves from the standardized approach to
the foundation IRB approach to the
advanced IRB approach
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Adjustments for Collateral
l Two approaches
l Simple approach: risk weight of counterparty
replaced by risk weight of collateral
l Comprehensive approach: exposure adjusted
upwards to allow to possible increases; value
of collateral adjusted downward to allow for
possible decreases; new exposure equals
excess of adjusted exposure over adjusted
collateral; counterparty risk weight applied to
the new exposure
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Guarantees
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Operational Risk Capital
l Basic Indicator Approach: 15% of gross
income
l Standardized Approach: different
multiplicative factor for gross income
arising from each business line
l Internal Measurement Approach: assess
99.9% worst case loss over one year.
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Supervisory Review Changes
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Market Discipline
l Banks will be required to disclose
l Scope and application of Basel framework
l Nature of capital held
l Regulatory capital requirements
l Nature of institution’s risk exposures
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Possible Revisions to Basel II
l Incremental risk charge (credit items in
trading book treated in the same way as if
they were in banking book)
l Stressed VaR (takes account of
movements in market variables during a
one-year period of significant losses in
calculating market risk capital)
l Movement away from self-regulation
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Solvency II
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Solvency II continued
l Internal models vs standardized approach
l One year 99.5% confidence for internal models
l Capital charge for investment risk, underwriting risk, and
operational risk
l Three types of capital
Risk Management and Financial Institutions 2e, Chapter 11, Copyright © John C. Hull 2009
Volatility
Chapter 9
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 38
Definition of Volatility
l Suppose that Si is the value of a variable on
day i. The volatility per day is the standard
deviation of ln(Si /Si-1)
l Normally days when markets are closed are
ignored in volatility calculations (see Business
Snapshot 9.1, page 177)
l The volatility per year is 252 times the daily
volatility
l Variance rate is the square of volatility
l Realized vs. Implied Vs. Conditional Volatility
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 39
Implied Volatilities
l Of the variables needed to price an option
the one that cannot be observed directly is
volatility
l We can therefore imply volatilities from
market prices and vice versa
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 40
VIX Index: A Measure of the Implied
Volatility of the S&P 500 (Figure 9.1, page
178)
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 41
Global Risk Management
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 42
Are Daily Changes in Exchange Rates
Normally Distributed? Table 9.2, page 181
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 43
Heavy Tails
l Daily exchange rate changes are not normally
distributed
l The distribution has heavier tails than the normal
distribution
l It is more peaked than the normal distribution
l This means that small changes and large
changes are more likely than the normal
distribution would suggest
l Many market variables have this property,
known as excess kurtosis
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 44
Normal and Heavy-Tailed
Distribution
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 45
Alternatives to Normal Distributions:
The Power Law (See page 182)
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 46
Log-Log Test for Exchange Rate
Data
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 47
Standard Approach to Estimating
Volatility
l Define sn as the volatility per day between
day n-1 and day n, as estimated at end of day
n-1
l Define Si as the value of market variable at
end of day i
l Define ui= ln(Si/Si-1)
m
1
s n2 = å
m - 1 i =1
( un -i - u ) 2
1 m
u = å un -i
m i =1
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 48
Simplifications Usually Made in
Risk Management
l Define ui as (Si−Si-1)/Si-1
l Assume that the mean value of ui is zero
l Replace m-1 by m
This gives
1 m 2
s = åi =1 un -i
2
n
m
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 49
Weighting Scheme
Instead of assigning equal weights to the
observations we can set
s = å i =1 a i u
2 m 2
n n -i
where
m
åa
i =1
i =1
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 50
ARCH(m) Model
In an ARCH(m) model we also assign
some weight to the long-run variance rate,
VL:
s = gVL + åi =1 a i u n2-i
2 m
n
where
m
g + å ai = 1
i =1
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 51
EWMA Model (page 186)
s = ls
2
n
2
n -1 + (1 - l)u 2
n -1
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 52
Attractions of EWMA
l Relatively little data needs to be stored
l We need only remember the current
estimate of the variance rate and the most
recent observation on the market variable
l Tracks volatility changes
l RiskMetrics uses l = 0.94 for daily
volatility forecasting
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 53
GARCH (1,1), page 188
In GARCH (1,1) we assign some weight to
the long-run average variance rate
s = gVL + au
2
n
2
n -1 + bs 2
n -1
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 54
GARCH (1,1) continued
Setting w = gVL the GARCH (1,1) model is
s = w + au
2
n
2
n -1 + bs 2
n -1
and
w
VL =
1- a - b
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 55
Example
l Suppose
s = 0.000002 + 013
2
n . u 2
n -1 + 0.86s 2
n -1
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 56
Example continued
l Suppose that the current estimate of the
volatility is 1.6% per day and the most
recent percentage change in the market
variable is 1%.
l The new variance rate is
0.000002 + 013
. ´ 0.0001 + 0.86 ´ 0.000256 = 0.00023336
The new volatility is 1.53% per day
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 57
GARCH (p,q)
p q
s = w + å ai u
2
n
2
n -i + å b js 2
n- j
i =1 j =1
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 58
Other Models
l Many other GARCH models have been
proposed
l For example, we can design a GARCH
models so that the weight given to ui2
depends on whether ui is positive or
negative
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 59
Variance Targeting
l One way of implementing GARCH(1,1)
that increases stability is by using variance
targeting
l We set the long-run average volatility
equal to the sample variance
l Only two other parameters then have to be
estimated
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 60
Maximum Likelihood Methods
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 61
Example 1 (page 190)
l We observe that a certain event happens
one time in ten trials. What is our estimate
of the proportion of the time, p, that it
happens?
l The probability of the outcome is
p(1 - p) 9
n
é u ù 2
å ê- ln(vi ) - ú
i =1 ë vi û
i
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 64
Calculations for Yen Exchange
Rate Data (Table 9.4, page 192)
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 65
Daily Volatility of Yen: 1988-1997
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 66
Forecasting Future Volatility
(Equation 9.14, page 195)
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 67
Forecasting Future Volatility cont
Risk Management and Financial Institutions 2e, Chapter 9, Copyright © John C. Hull 2009 69
Risk Management
and Financial Returns
Elements of Financial Risk Management
Chapter 1
Peter F. Christoffersen
0,10
0,05
0,00
-0,05
-0,10
-0,15
0 10 20 30 40 50 60 70 80 90 100
Lag Order
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Stylized fact 2
l The unconditional distribution of daily returns have
fatter tail than the normal distribution.
l Fig.1.2 shows a histogram of the daily S&P500 return
data with the normal distribution imposed.
l Notice how the histogram has longer and fatter tails,
in particular in the left side, and how it is more
peaked around zero than the normal distribution.
l Fatter tails mean a higher probability of large losses
than the normal distribution would suggest.
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Figure 1.2
Histogram of Daily S&P 500 Returns and the Normal
Distribution
Jan 1, 2001 - Dec 31, 2010
0,3000
0,2500
Normal Frequency
Probability Distribution
0,2000
0,1500
0,1000
0,0500
0,0000
-0,07 -0,06 -0,05 -0,04 -0,03 -0,02 -0,01 0,00 0,01 0,02 0,03 0,04 0,05 0,06 0,07
0,40
0,35
0,30
0,25
0,20
0,15
0,10
0,05
0,00
0 10 20 30 40 50 60 70 80 90 100
Lag Order
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Stylized fact 6
l Equity and equity indices display negative correlation
between variance and returns.
l This often termed the leveraged effect, arising from
the fact that a drop in stock price will increase the
leverage of the firm as long as debt stays constant.
l This increase in leverage might explain the increase
variance associated with the price drop. We will
model the leverage effect in Chapters 4 and 5.
l Also note that for this course we will use VaR based
on log returns defined as
16
14
Return Distribution
12
10
0
-0,08 -0,05 -0,03 0,00 0,03 0,05 0,08
Portfolio Return
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Figure 1.4
Value at Risk (VaR) from the Normal Distribution
Return Probability Distribution
1
0,9
0,8
Cumulative Distribution
0,7
0,6
0,5
0,4
0,3
0,1
0
-0,08 -0,05 -0,03 0,00 0,03 0,05 0,08
Portfolio Return
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Introducing the VaR risk
measure
l Consider a portfolio whose value consists of 40
shares in Microsoft (MS) and 50 shares in GE.
l To calculate VaR for the portfolio, collect historical
share price data for MS and GE and construct the
historical portfolio pseudo returns
0,120
0,100
0,080
VaR
0,060
0,040
0,020
0,000
Sep 01
Sep 02
Sep 03
Sep 04
Sep 05
Sep 06
Sep 07
Sep 08
Sep 09
Sep 10
Jan 01
Jan 02
Jan 03
Jan 04
Jan 05
Jan 06
Jan 07
Jan 08
Jan 09
Jan 10
Mai 01
Mai 02
Mai 03
Mai 04
Mai 05
Mai 06
Mai 07
Mai 08
Mai 09
Mai 10
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Drawbacks
l ofignored
Extreme losses are VaR - The VaR number only tells
us that 1% of the time we will get a return below the
reported VaR number, but it says nothing about what
will happen in those 1% worst cases.
l VaR assumes that the portfolio is constant across the
next K days, which is unrealistic in many cases when K
is larger than a day or a week.
l Finally, it may not be clear how K and p should be
chosen.
0,10
0,05
0,00
-0,05
-0,10
-0,15
0 10 20 30 40 50 60 70 80 90 100
Lag Order
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Stylized fact 2
l The unconditional distribution of daily returns have
fatter tail than the normal distribution.
l Fig.1.2 shows a histogram of the daily S&P500 return
data with the normal distribution imposed.
l Notice how the histogram has longer and fatter tails,
in particular in the left side, and how it is more
peaked around zero than the normal distribution.
l Fatter tails mean a higher probability of large losses
than the normal distribution would suggest.
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Figure 1.2
Histogram of Daily S&P 500 Returns and the Normal
Distribution
Jan 1, 2001 - Dec 31, 2010
0,3000
0,2500
Normal Frequency
Probability Distribution
0,2000
0,1500
0,1000
0,0500
0,0000
-0,07 -0,06 -0,05 -0,04 -0,03 -0,02 -0,01 0,00 0,01 0,02 0,03 0,04 0,05 0,06 0,07
0,40
0,35
0,30
0,25
0,20
0,15
0,10
0,05
0,00
0 10 20 30 40 50 60 70 80 90 100
Lag Order
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Stylized fact 6
l Equity and equity indices display negative correlation
between variance and returns.
l This often termed the leveraged effect, arising from
the fact that a drop in stock price will increase the
leverage of the firm as long as debt stays constant.
l This increase in leverage might explain the increase
variance associated with the price drop. We will
model the leverage effect in Chapters 4 and 5.
l Also note that for this course we will use VaR based
on log returns defined as
16
14
Return Distribution
12
10
0
-0,08 -0,05 -0,03 0,00 0,03 0,05 0,08
Portfolio Return
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Figure 1.4
Value at Risk (VaR) from the Normal Distribution
Return Probability Distribution
1
0,9
0,8
Cumulative Distribution
0,7
0,6
0,5
0,4
0,3
0,1
0
-0,08 -0,05 -0,03 0,00 0,03 0,05 0,08
Portfolio Return
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Introducing the VaR risk
measure
l Consider a portfolio whose value consists of 40
shares in Microsoft (MS) and 50 shares in GE.
l To calculate VaR for the portfolio, collect historical
share price data for MS and GE and construct the
historical portfolio pseudo returns
0,120
0,100
0,080
VaR
0,060
0,040
0,020
0,000
Sep 01
Sep 02
Sep 03
Sep 04
Sep 05
Sep 06
Sep 07
Sep 08
Sep 09
Sep 10
Jan 01
Jan 02
Jan 03
Jan 04
Jan 05
Jan 06
Jan 07
Jan 08
Jan 09
Jan 10
Mai 01
Mai 02
Mai 03
Mai 04
Mai 05
Mai 06
Mai 07
Mai 08
Mai 09
Mai 10
Elements of Financial Risk
Management Second Edition ©
2012 by Peter Christoffersen
Drawbacks
l ofignored
Extreme losses are VaR - The VaR number only tells
us that 1% of the time we will get a return below the
reported VaR number, but it says nothing about what
will happen in those 1% worst cases.
l VaR assumes that the portfolio is constant across the
next K days, which is unrealistic in many cases when K
is larger than a day or a week.
l Finally, it may not be clear how K and p should be
chosen.
168
Elements of Financial
Risk Management
Overview
Objectives
l Introduce the most commonly used method
for computing VaR, namely Historical
Simulation and discuss the pros and cons
of this method.
171Elements of Financial
Risk Management
Defining Historical Simulation
l This is a pseudo value because the units of
each asset held typically changes over time.
The pseudo log return can now be defined
as
• Consider the availability of a past sequence of m
daily hypothetical portfolio returns, calculated
using past prices of the underlying assets of the
portfolio, but using today’s portfolio weights, call
it {RPF,t+1-t}mt=1
172Elements of Financial
Risk Management
Defining Historical Simulation
l Distribution of RPF,t+1 is captured by the
histogram of {RPF,t+1-t}mt=1
l The VaR with coverage rate, p is calculated as
100pth percentile of the sequence of past
portfolio returns.
Cons
l It is very easy to implement. No numerical
optimization has to be performed.
l It is model-free. It does not rely on any
particular parametric model such as a
RiskMetrics
174Elements of Financial
Risk Management
model.
Issues with model free nature of HS
How large should m be?
l If m is too large, then the most recent
observations will carry very little weight, and
the VaR will tend to look very smooth over
time.
l If m is too small, then the sample may not
include enough large losses to enable the risk
manager to calculate VaR with any precision.
l To calculate 1% VaRs with any degree of
precision
175Elements of Financial
for the next 10 days, HS technique
Riskneeds
Managementa large m value
Figure 2.1:
VaRs from HS with 250 and 1,000 Return Days
Jul 1, 2008 - Dec 31, 2010
0,0900
0,0800
Historical Simulation VaR
0,0700
0,0600
0,0500
0,0400
HS-VaR(250)
HS-VaR(1000)
0,0300
0,0200
Jun 08 Jul 08 Sep 08 Okt 08 Dez 08 Feb 09 Mär 09 Mai 09 Jul 09 Aug 09 Okt 09 Dez 09 Jan 10
176Elements of Financial
Risk Management
Weighted Historical Simulation
l WHS relieves the tension in the choice of m
l It assigns relatively more weight to the most recent
observations and relatively less weight to the
returns further in the past
l It is implemented as follows –
l Sample of m past hypothetical returns, {RPF,t+1-
} m
t t=1 is assigned probability weights
declining exponentially through the past as follows
177Elements of Financial
Risk Management
Weighted Historical Simulation
l Today’s observation is assigned the weight h1 =
(1- h) / (1- hm)
l ht goes to zero as t gets large, and that the weights
ht for t = 1,2,...,m sum to 1
l Typical value for h is between 0.95 and 0.99
l The observations along with their assigned
weights are sorted in ascending order.
l The 100p% VaR is calculated by accumulating
the weights of the ascending returns until
100p% is reached.
178Elements of Financial
Risk Management
Pros and Cons of WHS
Pros
181Elements of Financial
Risk Management
Figure 2.2 A:
Historical Simulation VaR and Daily Losses from
Long S&P500 Position, October 1987
25%
15%
HS VaR and Loss
10%
5%
0%
-5%
-10%
-15%
01-Okt 06-Okt 11-Okt 16-Okt 21-Okt 26-Okt 31-Okt
Loss Date
182Elements of Financial
Risk Management
Figure 2.2 B:
Historical Simulation VaR and Daily Losses from
Short S&P500 Position, October 1987
25%
15%
HS VaR and Loss
10%
5%
0%
-5%
-10%
-15%
01-Okt 06-Okt 11-Okt 16-Okt 21-Okt 26-Okt 31-Okt
Loss Date
183Elements of Financial
Risk Management
Figure 2.3 A:
Historical Simulation VaR and Daily Losses from
Short S&P500 Position, October 1987
15%
5%
HS VaR and Loss
0%
-5%
-10%
-15%
-20%
-25%
01-Okt 06-Okt 11-Okt 16-Okt 21-Okt 26-Okt 31-Okt
Loss Date
184Elements of Financial
Risk Management
Figure 2.3 B:
Weighted Historical Simulation VaR and Daily Losses
from Short S&P500 Position, October 1987
15%
5%
WHS VaR and Loss
0%
-5%
-10%
-15%
-20%
-25%
01-Okt 06-Okt 11-Okt 16-Okt 21-Okt 26-Okt 31-Okt
Loss Date
185Elements of Financial
Risk Management
Evidence from the 2008-2009
Crisis
l We consider the daily closing prices for a
total return index of the S&P 500 starting in
July 2008 and ending in December 2009.
S&P500
2.000
1.800
S&P500-Closing Price
1.600
1.400
1.200
1.000
Jul 08 Sep 08 Nov 08 Jan 09 Mär 09 Mai 09 Jul 09 Sep 09 Nov 09
187Elements of Financial
Risk Management
Evidence from the 2008-2009 Crisis
l The 10-day 1% HS VaR is computed from the
1-day VaR by simply multiplying it by
188Elements of Financial
Risk Management
Evidence from the 2008-2009 Crisis
l where the variance dynamics are driven by
RiskMetrics HS
35%
30%
Value at Risk (VaR)
25%
20%
15%
10%
5%
0%
Jul 08 Aug 08 Sep 08 Okt 08 Nov 08 Dez 08 Jan 09 Feb 09Mär 09 Apr 09 Mai 09 Jun 09 Jul 09 Aug 09 Sep 09 Okt 09 Nov 09 Dez 09
190Elements of Financial
Risk Management
Evidence from the 2008-2009 Crisis
l The units in figure above refer to the least
percent of capital that would be lost over the
next 10 days in the 1% worst outcomes.
192Elements of Financial
Risk Management
Figure 2.6: Cumulative P/L from Traders with
HS and RM VaRs 150.000
0
Cumulati ve P/L
-50.000
-100.000
-150.000
-200.000
-250.000
-300.000
-350.000
-400.000
Jul Jul Aug Aug Sep Okt Okt Nov Dez Dez Jan Feb Mär Mär Apr Apr Mai Jun Jun Jul Aug Aug Sep Okt Okt Nov Dez Dez
08 08 08 08 08 08 08 08 08 08 09 09 09 09 09 09 09 09 09 09 09 09 09 09 09 09 09 09
193Elements of Financial
Risk Management
Evidence from the 2008-2009 Crisis
Performance difference between HS and RM
VaRs
l We can write:
195Elements of Financial
Risk Management
The Probability of Breaching the HS VaR
l Simulate 1,250 return observations from above
equation
l Starting on day 251, compute each day the 1-
day, 1% VaR using Historical Simulation
l Compute the true probability that we will
observe a loss larger than the HS VaR we have
computed
l This is the probability of a VaR breach
196Elements of Financial
Risk Management
Figure 2.7: Actual Probability of Loosing More
than the 1% HS VaR When Returns Have
Dynamics Variance
0,18
0,16
0,14
Probability of VaR Breach
0,12
0,10
0,08
0,06
0,04
0,02
0,00
1 101 201 301 401 501 601 701 801 901
197Elements of Financial
Risk Management
The Probability of Breaching the HS VaR
l where is the cumulative density function
for a standard normal random variable.
l If the HS VaR model had been accurate then
this plot should show a roughly flat line at
1%
l Here we see numbers as high as 16% and
numbers very close to 0%
l The HS VaR will overestimate risk when true
market volatility is low, which will generate a
low probability of a VaR breach
HS will underestimate risk when true
198Elements of Financial
lManagement
Risk
VaR with Extreme Coverage Rates
l The tail of the portfolio return distribution
conveys information about the future losses.
199Elements of Financial
When using HS with a 250-day sample it is not
Risk Management
l
Figure 2.8: Relative Difference between Non-
Normal (Excess Kurtosis=3) and Normal VaR
80
70
60
Relative VaR Difference
50
40
30
20
10
0
0 0,005 0,01 0,015 0,02 0,025
VaR Coverage Rate, p
200Elements of Financial
Risk Management
VaR with Extreme Coverage Rates
l Note that (from the above figure) as p gets
close to zero the nonnormal VaR gets much
larger than the normal VaR
201Elements of Financial
Risk Management
Expected Shortfall
l VaR is concerned only with the percentage of
losses that exceed the VaR and not the magnitude
of these losses.
l Expected Shortfall (ES), or TailVaR accounts for
the magnitude of large losses as well as their
probability of occurring
l Mathematically ES is defined as
202Elements of Financial
Risk Management
Expected Shortfall
l The negative signs in front of the expectation
and the VaR are needed because the ES and
the VaR are defined as positive numbers
l The ES tells us the expected value of
tomorrow’s loss, conditional on it being worse
than the VaR
l The Expected Shortfall computes the average
of the tail outcomes weighted by their
probabilities
l ES tells us the expected loss given that we
203Elements of Financial
actually get a loss from the 1% tail
Risk Management
Expected Shortfall
l To compute ES we need the distribution of a
normal variable conditional on it being below
the VaR
204Elements of Financial
Risk Management
Expected Shortfall
l f(•) denotes the density function and F(•)
the cumulative density function of the
standard normal distribution
l In the normal distribution case ES can be
derived as
205Elements of Financial
Risk Management
Expected Shortfall
l In the normal case we know that
• Thus, we have
206Elements of Financial
Risk Management
Expected Shortfall
l For example, when p =0.01, we have
and the relative difference is then
• From the below figure, the blue line shows that when
excess kurtosis is zero, the relative difference between
the ES and VaR is 15%
207Elements of Financial
Risk Management
Expected Shortfall
l The blue line also shows that for moderately
large values of excess kurtosis, the relative
difference between ES and VaR is above 30%
l From the figure, the relative difference
between VaR and ES is larger when p is larger
and thus further from zero
l When p is close to zero VaR and ES will both
capture the fat tails in the distribution
l When p is far from zero, only the ES will
capture the fat tails in the return distribution
208Elements of Financial
Risk Management
Figure 2.9: ES vs VaR as a Function of Kurtosis
50
p=1% p=5%
45
40
35
(ES - VaR ) / ES
30
25
20
15
10
0
0 1 2 3 4 5 6
Excess Kurtosis
209Elements of Financial
Risk Management
Summary
Chapter 8
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 211
The Question Being Asked in VaR
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 212
Value at Risk (VaR)
l “VaR measures the worst expected loss over a given time interval
l under normal market conditions at a given confidence level.”
l - Jorion (1997)
l
l “Value at Risk is an estimate, with a given degree of confidence, of how
much one can lose from one’s portfolio over a given time horizon.”
l - Wilmott (1998)
l The standard deviation of the asset price (S) over a period t is:
l S (s t.5) = S ( sÖ t )
l For example, let
l S = $ 100.00
l s = 40%
l t = 1 week = 1/52
l similarly,
l daily s.d. = 100(0.40)(1/252).5 = 40(0.063) = $ 2.52
l monthly s.d. = 0.40(0.40)(1/12).5 = 40(0.289) = $ 11.55
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 214
Confidence levels and the inverse distribution function
`
Let VaR = - S (sÖt) N (1 - confidence level)
(for long position in underlying asset)
`
where N (x%) = inverse cumulative distribution function
for the standard normal
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 216
Advantages of VaR
l It captures an important aspect of risk
in a single number
l It is easy to understand
l It asks the simple question: “How bad can
things get?”
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 217
Example 8.1 (page 159)
l The gain from a portfolio during six month
is normally distributed with mean $2
million and standard deviation $10 million
l The 1% point of the distribution of gains is
2−2.33×10 or − $21.3 million
l The VaR for the portfolio with a six month
time horizon and a 99% confidence level is
$21.3 million.
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 218
Example 8.2 (page 159)
l All outcomes between a loss of $50 million
and a gain of $50 million are equally likely
for a one-year project
l The VaR for a one-year time horizon and a
99% confidence level is $49 million
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 219
Examples 8.3 and 8.4 (page 160)
l 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
l The VaR with a 99% confidence level is $4
million
l What if the confidence level is 99.9%?
l What if it is 99.5%?
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 220
Cumulative Loss Distribution for
Examples 8.3 and 8.4 (Figure 8.3, page 160)
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 221
VaR vs. Expected Shortfall
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 222
Distributions with the Same VaR but
Different Expected Shortfalls
VaR
VaR
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 223
Coherent Risk Measures (page 162)
l Define a coherent risk measure as the amount of
cash that has to be added to a portfolio to make its
risk acceptable
l Properties of coherent risk measure
l If one portfolio always produces a worse outcome than
another its risk measure should be greater
l If we add an amount of cash K to a portfolio its risk measure
should go down by K
l Changing the size of a portfolio by l should result in the risk
measure being multiplied by l
l 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
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 224
VaR vs Expected Shortfall
l VaR satisfies the first three conditions but
not the fourth one
l Expected shortfall satisfies all four
conditions.
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 225
Example 8.5 and 8.7
l 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
l What is the 97.5% VaR for each project?
l What is the 97.5% expected shortfall for each project?
l What is the 97.5% VaR for the portfolio?
l What is the 97.5% expected shortfall for the portfolio?
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 226
Examples 8.6 and 8.8
l Two $10 million one-year loans each has a 1.25%
chance of defaulting. All recoveries between 0 and 100%
are equally likely. If there is no default the loan leads to a
profit of $0.2 million. If one loan defaults it is certain that
the other one will not default.
l What is the 99% VaR and expected shortfall of each
project
l What is the 99% VaR and expected shortfall for the
portfolio
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 227
Spectral Risk Measures
l A spectral risk measure assigns weights to
quantiles of the loss distribution
l VaR assigns all weight to Xth quantile of
the loss distribution
l Expected shortfall assigns equal weight to
all quantiles greater than the Xth quantile
l For a coherent risk measure weights must
be a non-decreasing function of the
quantiles
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 228
Normal Distribution Assumption
l The simplest assumption is that daily
gains/losses are normally distributed and
independent with mean zero
l It is then easy to calculate VaR from the
standard deviation (1-day VaR=2.33s)
l The T-day VaR equals T times the one-day
VaR
l Regulators allow banks to calculate the 10 day
VaR as 10 times the one-day VaR
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 229
Independence Assumption in VaR
Calculations (Equation 8.3, page 166)
l When daily changes in a portfolio are
identically distributed and independent the
variance over T days is T times the
variance over one day
l When there is autocorrelation equal to r
the multiplier is increased from T to
T + 2(T - 1)r + 2(T - 2)r 2 + 2(T - 3)r 3 + ! 2r T -1
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 230
Impact of Autocorrelation: Ratio of N-day
VaR to 1-day VaR (Table 8.1, page 204)
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 231
Choice of VaR Parameters
l Time horizon should depend on how quickly
portfolio can be unwound. Bank regulators in
effect use 1-day for market risk and 1-year for
credit/operational risk. Fund managers often
use one month
l Confidence level depends on objectives.
Regulators use 99% for market risk and 99.9%
for credit/operational risk.
l A bank wanting to maintain a AA credit rating will
often use confidence levels as high as 99.97%
for internal calculations.
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 232
VaR Measures for a Portfolio where an
amount xi is invested in the ith component
of the portfolio (page 168-169)
l Marginal VaR: ¶VaR
¶xi
¶VaR
l Component VaR: xi
¶xi
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 233
Properties of Component VaR
l The component VaR is approximately the
same as the incremental VaR
l The total VaR is the sum of the component
VaR’s (Euler’s theorem)
l The component VaR therefore provides a
sensible way of allocating VaR to different
activities
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 234
Back-testing (page 169-171)
l Back-testing a VaR calculation methodology involves
looking at how often exceptions (loss > VaR) occur
l 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
l Suppose that the theoretical probability of an exception
is p (=1−X). The probability of m or more exceptions in n
days is
n
n!
å
k = m k!( n - k )!
p k
(1 - p ) n-k
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 235
Bunching
l Bunching occurs when exceptions are not
evenly spread throughout the back testing
period
l Statistical tests for bunching have been
developed by Christoffersen (See page 171)
Risk Management and Financial Institutions 2e, Chapter 8, Copyright © John C. Hull 2009 236
Correlations and
Copulas
Chapter 11
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 237
Correlation and Covariance
l The coefficient of correlation between two
variables V1 and V2 is defined as
l The covariance is
E(V1V2)−E(V1 )E(V2)
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 238
Independence
l V1 and V2 are independent if the
knowledge of one does not affect the
probability distribution for the other
f (V2 V1 = x ) = f (V2 )
where f(.) denotes the probability density
function
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 239
Independence is Not the Same as
Zero Correlation
l Suppose V1 = –1, 0, or +1 (equally
likely)
l If V1 = -1 or V1 = +1 then V2 = 1
l If V1 = 0 then V2 = 0
V2 is clearly dependent on V1 (and vice
versa) but the coefficient of correlation
is zero
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 240
Types of Dependence (Figure 11.1, page 235)
E(Y) E(Y)
X X
(a) (b)
E(Y)
(c)
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 241
Monitoring Correlation Between
Two Variables X and Y
Define xi=(Xi−Xi-1)/Xi-1 and yi=(Yi−Yi-1)/Yi-1
Also
varx,n: daily variance of X calculated on day n-1
vary,n: daily variance of Y calculated on day n-1
covn: covariance calculated on day n-1
The correlation is
cov n
varx ,n vary ,n
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 242
Covariance
l The covariance on day n is
E(xnyn)−E(xn)E(yn)
l It is usually approximated as E(xnyn)
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 243
Monitoring Correlation continued
EWMA:
cov n = l cov n -1 + (1 - l) xn -1 yn -1
GARCH(1,1)
cov n = w + axn -1 yn -1 + b cov n -1
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 244
Positive Finite Definite Condition
A variance-covariance matrix, W, is
internally consistent if the positive semi-
definite condition
wTWw ≥ 0
holds for all vectors w
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 245
Example
The variance covariance matrix
æ 1 0 0.9ö
ç ÷
ç 0 1 0.9÷
ç ÷
è 0.9 0.9 1ø
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 246
V1 and V2 Bivariate Normal
l Conditional on the value of V1, V2 is normal with
mean
V1 - µ1
µ 2 + rs 2
s1
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 248
Generating Random Samples for
Monte Carlo Simulation (pages 239-240)
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 249
Factor Models (page 240)
l When there are N variables, Vi (i = 1,
2,..N), in a multivariate normal distribution
there are N(N−1)/2 correlations
l We can reduce the number of correlation
parameters that have to be estimated with
a factor model
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 250
One-Factor Model continued
l If Ui have standard normal distributions
we can set
U i = ai F + 1 - ai2 Z i
where the common factor F and the
idiosyncratic component Zi have
independent standard normal
distributions
l Correlation between Ui and Uj is ai aj
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 251
Gaussian Copula Models:
Creating a correlation structure for variables that are not
normally distributed
-0.2 0 0.2 0.4 0.6 0.8 1 1.2 -0.2 0 0.2 0.4 0.6 0.8 1 1.2
V1 V2
One-to-one
mappings
-6 -4 -2 0 2 4 6 -6 -4 -2 0 2 4 6
U2
U1
Correlation
Assumption
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 253
Example (page 241)
V1 V2
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 254
V1 Mapping to U1
V1 Percentile U1
0.2 20 -0.84
0.4 55 0.13
0.6 80 0.84
0.8 95 1.64
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 255
V2 Mapping to U2
V2 Percentile U2
0.2 8 −1.41
0.4 32 −0.47
0.6 68 0.47
0.8 92 1.41
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 256
Example of Calculation of Joint
Cumulative Distribution
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 257
Other Copulas
l Instead of a bivariate normal distribution
for U1 and U2 we can assume any other
joint distribution
l One possibility is the bivariate Student t
distribution
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 258
5000 Random Samples from the
Bivariate Normal
0
-5 -4 -3 -2 -1 0 1 2 3 4 5
-1
-2
-3
-4
-5
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 259
5000 Random Samples from the
Bivariate Student t
10
0
-10 -5 0 5 10
-5
-10
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 260
Multivariate Gaussian Copula
l We can similarly define a correlation
structure between V1, V2,…Vn
l We transform each variable Vi to a new
variable Ui that has a standard normal
distribution on a “percentile-to-percentile”
basis.
l The U’s are assumed to have a
multivariate normal distribution
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 261
Factor Copula Model
In a factor copula model the correlation
structure between the U’s is generated by
assuming one or more factors.
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 262
Credit Default Correlation
l The credit default correlation between two
companies is a measure of their tendency
to default at about the same time
l Default correlation is important in risk
management when analyzing the benefits
of credit risk diversification
l It is also important in the valuation of some
credit derivatives
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 263
Model for Loan Portfolio
l We map the time to default for company i, Ti, to a
new variable Ui and assume
U i = aF + 1- a Z i 2
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 264
Vasicek’s Model for Loan Portfolio
l Suppose the Bank has a large no. of Loans where PD is 1% but will
change in each year. Suppose Ti is the time firm defaults. All firms has
the same Cum. Prob. Dist, for the time to default Qi.
l The Gaussian copula method can be used to define correlation structure
between Ti’s.
l We map the time to default for company i, Ti, to a new variable Ui N(0,1)
and assume
U i = aF + 1- a Z i 2
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 265
Vasicek’s Model for Loan Portfolio
l
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 266
Analysis
l To analyze the model we
l Calculate the probability that, conditional on the value
of F, Ui is less than some value U
l This is the same as the probability that Ti is less that T
where T and U are the same percentiles of their
distributions
This leads to
é N -1 [PD] - r F ù
Prob(Ti < T F ) = N ê ú
êë 1- r úû
where PD is the probability of default in time T
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 267
The Model continued
l The X% worst case value of F is N-1(X)
l The worst case default rate during time T with a
confidence level of X is therefore
æ N -1[Q(T )] + r N -1 ( X ) ö
WCDR(T,X) = N ç ÷
ç 1- r ÷
è ø
l The VaR for this time horizon and confidence limit
is
VaR (T , X ) = L ´ LGD ´ WCDR (T , X )
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 268
Gordy’s Result
l In a large portfolio of M loans where each
loan is small in relation to the size of the
portfolio it is approximately true that
M
VaR (T , X ) = å Li ´ LGDi ´ WCDR i (T , X )
i =1
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 269
Estimating PD and r
l We can use data on default rates in
conjunction with maximum likelihood
methods
l The probability density function for the
default rate is
ì é ü
1 - r N (DR ) - N (PD) ù ï
2
1- r ï 1 ê -1 æ -1 -1
ö
g (DR ) = expí ( N (DR )) - çç
2
÷ úý
÷ ú
r ïî 2 êë è r ø û ïþ
Risk Management and Financial Institutions 3e, Chapter 11, Copyright © John C. Hull 2012 270
Value at Risk
Chapter 9
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 271
The Question Being Asked in VaR
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 272
Value at Risk - Definitions
l VaR measures the worst expected loss over a given time interval
l under normal market conditions at a given confidence level.”
l - Jorion (1997)
l
l “Value at Risk is an estimate, with a given degree of confidence, of how
much one can lose from one’s portfolio over a given time horizon.”
l - Wilmott (1998)
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 274
Advantages of VaR
l It captures an important aspect of risk
in a single number
l It is easy to understand
l It asks the simple question: “How bad can
things get?”
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 275
Example 9.1 (page 185)
l The gain from a portfolio during six month
is normally distributed with mean $2
million and standard deviation $10 million
l The 1% point of the distribution of gains is
2−2.33×10 or − $21.3 million
l The VaR for the portfolio with a six month
time horizon and a 99% confidence level is
$21.3 million.
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 276
Example 9.2 (page 186)
l All outcomes between a loss of $50 million
and a gain of $50 million are equally likely
for a one-year project
l The VaR for a one-year time horizon and a
99% confidence level is $49 million
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 277
Examples 9.3 and 9.4 (page 186)
l 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
l The VaR with a 99% confidence level is $4
million
l What if the confidence level is 99.9%?
l What if it is 99.5%?
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 278
Cumulative Loss Distribution for
Examples 9.3 and 9.4 (Figure 9.3, page 186)
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 279
VaR vs. Expected Shortfall
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 280
Distributions with the Same VaR but
Different Expected Shortfalls
VaR
VaR
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 281
Coherent Risk Measures (page 188)
l Define a coherent risk measure as the amount of
cash that has to be added to a portfolio to make its
risk acceptable
l Properties of coherent risk measure
l If one portfolio always produces a worse outcome than
another its risk measure should be greater
l If we add an amount of cash K to a portfolio its risk measure
should go down by K
l Changing the size of a portfolio by l should result in the risk
measure being multiplied by l
l 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
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 282
VaR vs Expected Shortfall
l VaR satisfies the first three conditions but
not the fourth one
l Expected shortfall satisfies all four
conditions.
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 283
Example 9.5 and 9.7
l 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
l What is the 97.5% VaR for each project?
l What is the 97.5% expected shortfall for each project?
l What is the 97.5% VaR for the portfolio?
l What is the 97.5% expected shortfall for the portfolio?
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 284
Examples 9.6 and 9.8
l 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%
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 285
Spectral Risk Measures
l A spectral risk measure assigns weights to
quantiles of the loss distribution
l VaR assigns all weight to Xth quantile of
the loss distribution
l Expected shortfall assigns equal weight to
all quantiles greater than the Xth quantile
l For a coherent risk measure weights must
be a non-decreasing function of the
quantiles
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 286
Normal Distribution Assumption
l The simplest assumption is that daily
gains/losses are normally distributed and
independent with mean zero
l It is then easy to calculate VaR from the
standard deviation (1-day VaR=2.33s)
l The T-day VaR equals T times the one-day
VaR
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 287
Independence Assumption in VaR
Calculations (Equation 9.3, page 193)
l When daily changes in a portfolio are
identically distributed and independent the
variance over T days is T times the
variance over one day
l When there is autocorrelation equal to r
the multiplier is increased from T to
T + 2(T - 1)r + 2(T - 2)r 2 + 2(T - 3)r 3 + ! 2r T -1
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 288
Impact of Autocorrelation: Ratio of T-day
VaR to 1-day VaR (Table 9.1, page 193)
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 289
Choice of VaR Parameters
l Time horizon should depend on how quickly
portfolio can be unwound. Bank regulators in
effect use 1-day for market risk and 1-year for
credit/operational risk. Fund managers often
use one month
l Confidence level depends on objectives.
Regulators use 99% for market risk and 99.9%
for credit/operational risk.
l A bank wanting to maintain a AA credit rating
might use confidence levels as high as 99.97%
for internal calculations.
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 290
VaR Measures for a Portfolio where an
amount xi is invested in the ith component
of the portfolio (page 195-196)
l Marginal VaR: ¶VaR
¶xi
¶VaR
l Component VaR: xi
¶xi
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 291
Properties of Component VaR
l The component VaR is approximately the
same as the incremental VaR
l The total VaR is the sum of the component
VaR’s (Euler’s theorem)
l The component VaR therefore provides a
sensible way of allocating VaR to different
activities
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 292
Aggregating VaRs
An approximate approach that seems to works
well is
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 293
Back-testing (page 197-200)
l Back-testing a VaR calculation methodology involves
looking at how often exceptions (loss > VaR) occur
l 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
l Suppose that the theoretical probability of an exception
is p (=1−X). The probability of m or more exceptions in n
days is
n
n!
å
k = m k!( n - k )!
p k
(1 - p ) n-k
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 294
Bunching
l Bunching occurs when exceptions are not
evenly spread throughout the back testing
period
l Statistical tests for bunching have been
developed by Christoffersen (See page 171)
Risk Management and Financial Institutions 3e, Chapter 9, Copyright © John C. Hull 2012 295
Chapter 10
Properties of Stock
Options
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 297
Effect of Variables on Option
Pricing (Table 10.1, page 215)
Variable c p C P
S0 + − + −
K − + − +
T ? ? + +
s + + + +
r + − + −
D − + − +
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 298
American vs European Options
An American option is worth at least as
much as the corresponding European
option
C³c
P³p
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 299
Calls: An Arbitrage Opportunity?
l Suppose that
c=3 S0 = 20
T=1 r = 10%
K = 18 D=0
c ³ S0 –Ke -rT
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 301
Puts: An Arbitrage Opportunity?
l Suppose that
p= 1 S0 = 37
T = 0.5 r =5%
K = 40 D =0
p ³ Ke -rT–S0
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 303
Put-Call Parity: No Dividends
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 304
Values of Portfolios
ST > K ST < K
Portfolio A Call option ST − K 0
Zero-coupon bond K K
Total ST K
Portfolio C Put Option 0 K− ST
Share ST ST
Total ST K
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 305
The Put-Call Parity Result (Equation
10.6, page 222)
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 306
Arbitrage Opportunities
l Suppose that
c= 3 S0= 31
T = 0.25 r = 10%
K =30 D=0
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 308
An Extreme Situation
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 310
Bounds for European or American
Call Options (No Dividends)
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 311
Should Puts Be Exercised
Early ?
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 312
Bounds for European and American
Put Options (No Dividends)
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 313
The Impact of Dividends on Lower
Bounds to Option Prices
(Equations 10.8 and 10.9, page 229)
- rT
c ³ S 0 - D - Ke
- rT
p ³ D + Ke - S0
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 314
Extensions of Put-Call Parity
l American options; D = 0
S0 − K < C − P < S0 − Ke−rT
Equation 10.7 p. 224
Options, Futures, and Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 315