Backtesting Value-at-Risk: A Duration-Based Approach1

Peter Christoffersen2
McGill University, CIRANO and CIREQ

Denis Pelletier3
Université de Montreal, CIRANO and CIREQ

January 31, 2003

1 The first author acknowledges financial support from IFM2, FCAR, and SSHRC, and the second
author from FCAR and SSHRC. We are grateful for helpful comments from Frank Diebold, Jean-Marie
Dufour, Rob Engle, Eric Ghysels, James MacKinnon, Nour Meddahi, and Matt Pritsker. The usual
disclaimer applies.
2 Corresponding author. Faculty of Management, 1001 Sherbrooke Street West, Montreal, Quebec,

Canana H3A 1G5. Phone: (514) 398-2869. Fax: (514) 398-3876. Email: [Link]ffersen@[Link]
3 Department de Sciences Economiques, CP 6128, succursale Centre-Ville, Montreal, Quebec, H3C

3J7, Canada. Email: [Link]@[Link].

 Abstract

Financial risk model evaluation or backtesting is a key part of the internal model’s approach
to market risk management as laid out by the Basle Commitee on Banking Supervision (1996).
However, existing backtesting methods such as those developed in Christoffersen (1998), have
relatively small power in realistic small sample settings. Methods suggested in Berkowitz (2001)
fare better, but rely on information such as the shape of the left tail of the portfolio return
distribution, which is often not available. By far the most common risk measure is Value-at-Risk
(V aR), which is defined as a conditional quantile of the return distribution, and it says nothing
about the shape of the tail to the left of the quantile. Our contribution is the exploration of a new
tool for backtesting based on the duration of days between the violations of the V aR. The chief
insight is that if the V aR model is correctly specified for coverage rate, p, then the conditional
expected duration between violations should be a constant 1/p days. We suggest various ways
of testing this null hypothesis and we conduct a Monte Carlo analysis which compares the new
tests to those currently available. Our results show that in realistic situations, the duration based
tests have better power properties than the previously suggested tests. The size of the tests is
easily controlled using the Monte Carlo technique of Dufour (2000).
1 Motivation
Financial risk model evaluation or backtesting is a key part of the internal model’s approach to
market risk management as laid out by the Basle Committee on Banking Supervision (1996).
However, existing backtesting methods such as those developed in Christoffersen (1998), has
relatively small power in realistic small sample settings. Methods suggested in Berkowitz (2001)
fare better, but rely on information such as the shape of the left tail of the portfolio return
distribution, which is often not available. By far the most common risk measure is Value-at-Risk
(V aR), which is defined as a conditional quantile of the return distribution, and it says nothing
about the shape of the tail to the left of the quantile.
We will refer to an event where the ex-post portfolio loss exceeds the ex-ante V aR measure as a
violation. Of particular importance in backtesting is the clustering of violations. An institution’s
internal risk management team as well as external supervisors explicitly want to be able to detect
clustering in violations. Large losses which occur in rapid succession are more likely to lead to
disastrous events such as bankruptcy.
In the previous literature, due to the lack of real portfolio data, the evaluation of V aR
techniques were largely based on artificial portfolios. Examples in this tradition include Beder
(1995), Christoffersen, Hahn and Inoue (2001), Hendricks (1996), Kupiec (1995), Marshall and
Siegel (1997), and Pritsker (1997). But recently, Berkowitz and O’Brien (2002) have reported
on the performance of actual V aR forecasts from six large (and anonymous) U.S. commercial
banks.1 Figure 1 reproduces a picture from their paper which shows the V aR exceedences from
the six banks reported in standard deviations of the portfolio returns. Even though the banks
tend to be conservative—they have fewer than expected violations—the exceedences are large and
appear to be clustered in time and across banks. From the perspective of a regulator worried
about systemic risk, rejecting a particular bank’s risk model due to the clustering of violations
is particularly important if the violations also happen to be correlated across banks.
The detection of violation clustering is particularly important because of the widespread
reliance on V aRs calculated from the so-called Historical Simulation (HS) technique. In the
HS methodology, a sample of historical portfolio returns using current portfolio weights is first
constructed. The V aR is then simply calculated as the unconditional quantile from the historical
sample. The HS method thus largely ignores the last 20 years of academic research on conditional
asset return models. Time variability is only captured through the rolling historical sample. In
spite of forceful warnings, such as Pritsker (2001), the model-free nature of the HS technique
is viewed as a great benefit by many practitioners. The widespread use of HS the technique
motivates us to focus attention on backtesting V aRs calculated using this method.
1
Barone-Adesi, Giannopoulos and Vosper (2000) provides another example using real-life portfolio returns.

1
 While alternative methods for calculating portfolio measures such as the V aR have been inves-
tigated in for example Jorion (2000), and Christoffersen (2002), available methods for backtesting
are still relatively few. Our contribution is thus the exploration of a new tool for backtesting
based on the duration of days between the violations of the risk metric. The chief insight is
that if the V aR model is correctly specified for coverage rate, p, then the conditional expected
duration between violations should be a constant 1/p days. We suggest various ways of testing
this null hypothesis and we conduct a Monte Carlo analysis which compares the new tests to
those currently available. Our results show that in many realistic situations, the duration based
tests have better power properties than the previously suggested tests. The size of the tests is
easily controlled using the Monte Carlo testing approach of Dufour (2000). This procedure is
described in detail below.
We hasten to add that the sort of omnibus backtesting procedures suggested here are meant
as complements to—and not substitutes for—the statistical diagnostic tests carried out on various
aspects of the risk model in the model estimation stage. The tests suggested in this paper can
be viewed either as a final diagnostic for an internal model builder or alternatively as a feasible
diagnostic for an external model evaluator for whom only limited, aggregate portfolio information
is available.
Our paper is structured as follows: Section 2 outlines the previous first-order Markov tests,
Section 3 suggests the new duration-based tests, and Section 4 discusses details related to the
implementation of the tests. Section 5 contains Monte Carlo evidence on the performance of the
tests. Section 6 suggests various extensions to the analysis, and Section 7 concludes.

2 Extant Procedures for Backtesting Value-at-Risk

Consider a time series of daily ex-post portfolio returns, Rt , and a corresponding time series
of ex-ante Value-at-Risk forecasts, V aRt (p) with promised coverage rate p, such that ideally
Prt−1 (Rt < −V aRt (p)) = p. The negative sign arises from the convention of reporting the V aR
as a positive number.
Define the hit sequence of V aRt violations as
(
1, if Rt < −V aRt (p)
It =
0, else

Notice that the hit sequence appears to discard a large amount of information regarding the
size of violations etc. Recall, however, that the V aR forecast does not promise violations of a
certain magnitude, but rather only their conditional frequency, i.e. p. This is a major drawback
of the V aR risk measure which we will discuss below.

2
 Christoffersen (1998) tests the null hypothesis that

It ∼ i.i.d. Bernoulli(p)

against the alternative that

It ∼ i.i.d. Bernoulli(π)

and refers to this as the test of correct unconditional coverage (uc)

H0,uc : π = p

which is a test that on average the coverage is correct. The above test implicitly assumes that
the hits are independent an assumption which we now test explicitly. In order to test this
hypothesis an alternative is defined where the hit sequence follows a first order Markov sequence
with switching probability matrix
" #
1 − π01 π 01
Π=
1 − π11 π 11

where π ij is the probability of an i on day t − 1 being followed by a j on day t. The test of

independence (ind) is then
H0,ind : π 01 = π 11

Finally one can combine the two tests in a test of conditional coverage (cc)

H0,cc : π 01 = π 11 = p

The idea behind the Markov alternative is that clustered violations represent a signal of risk
model misspecification. Violation clustering is important as it implies repeated severe capital
losses to the institution which together could result in bankruptcy.
Notice however, that the Markov first-order alternative may have limited power against gen-
eral forms of clustering. The first point of this paper is to establish more general tests for
clustering which nevertheless only rely on information in the hit sequence.

3 Duration-Based Tests of Independence

The above tests are reasonably good at catching misspecified risk models when the temporal
dependence in the hit-sequence is of a simple first-order Markov structure. However we are
interested in developing tests which have power against more general forms of dependence but
which still rely only on estimating a few parameters.

3
 The intuition behind the duration-based tests suggested below is that the clustering of no-hit
durations will result in an excessive number of relatively short and relatively long durations,
corresponding to market turbulence and market calm respectively. Motivated by this intuition
we consider the duration of time (in days) between two V aR violations (i.e. the no-hit duration)
as
Di = ti − ti−1
where ti denotes the day of violation number i.2
Under the null hypothesis that the risk model is correctly specified, the no-hit duration should
have no memory and a mean duration of 1/p days. To verify the no memory property note that
under the null hypothesis we have the discrete probability distribution

Pr (D = 1) = p
Pr (D = 2) = (1 − p) p
Pr (D = 2) = (1 − p)2 p
...
Pr (D = d) = (1 − p)d−1 p.

A duration distribution is often best understood by its hazard function, which has the intuitive
definition of the probability of a getting a violation after D days given that we have gone D days
without a violation. The above probability distribution implies a flat discrete hazard function
as the following derivation shows
Pr (D = d)
λ (d) = P
1 − j<d Pr (D = d)
(1 − p)d−1 p
= P
1 − d−1 i
i=1 (1 − p) p
(1 − p)d−1 p
= P
1 − d−2 j
j=0 (1 − p) p
= p.

The only memory free (continuous)3 random distribution is the exponential, thus we have
that under the null the distribution of the no-hit durations should be

fexp (D; p) = p exp (−pD) .

2
For a general introduction to duration modeling, see Kiefer (1988) and Gourieroux (2000).
3
Notice that we use a continuous distribution even though we are counting time in days. This discreteness bias
will we acounted for in the Monte Carlo tests. The exponential distribution can also be viewed as the continuous
time limit of the above discrete time process. See Poirier (1995).

4
 In order to establish a statistical test for independence we must specify a (parsimonious)
alternative which allows for duration dependence. As a very simple case, consider the Weibull
distribution where
¡ ¢
fW (D; a, b) = ab bDb−1 exp −(aD)b .

The Weibull distribution has the advantage that the hazard function has a closed form rep-
resentation, namely
fW (D)
λW (D) ≡ = ab bDb−1
1 − FW (D)
where the exponential distribution appears as a special case with a flat hazard, when b = 1. The
Weibull will have a decreasing hazard function when b < 1, which corresponds to an excessive
number of very short durations (very volatile periods) and an excessive number of very long
durations (very tranquil periods). This could be evidence of misspecified volatility dynamics in
the risk model.
Due to the bankruptcy threat from VaR violation clustering the null hypothesis of indepen-
dence is of particular interest. We therefore want to explicitly test the null hypothesis

H0,ind : b = 1.

We could also use the Gamma distribution under the alternative hypothesis. The p.d.f. in
this case is
ab Db−1 exp (−aD)
fΓ (D; a, b) =
Γ (b)
which also nests the exponential when b = 1. In this case we therefore also have the independence
test null hypothesis as
H0,ind : b = 1.

The Gamma distribution does not have a closed-form solution for the hazard function, but
the first two moments are ab and ab2 respectively, so the notion of excess dispersion which is
defined as the variance over the squared expected value is simply 1b .
Note that the average duration in the exponential distribution is 1/p, and the variance of
durations is 1/p2 , thus the notion of excess dispersion is 1 in the exponential distribution.

3.1 A Conditional Duration Test

The above duration tests can potentially capture higher order dependence in the hit sequence by
simply testing the unconditional distribution of the durations. Dependence in the hit sequence
may show up as an excess of relatively long no-hit durations (quiet periods) and an excess of
relatively short no-hit durations, corresponding to violation clustering. However, in the above

5
tests, any information in the ordering of the durations is completely lost. The information
in the temporal ordering of no-hit durations could be captured using the framework of Engle
and Russel’s (1998) Exponential Autoregressive Conditional Duration (EACD) model. In the
EACD(1,0) model, the conditional expected duration takes the following form

Ei−1 [Di ] ≡ ψi = ω + αDi−1

with α ∈ [0, 1) . Assuming an underlying exponential density with mean equal to one, the condi-
tional distribution of the duration is
µ ¶
1 Di
fEACD (Di |ψi ) = exp −
ψi ψi
The null of independent no-hit durations would then correspond to

H0,ind : α = 0.

Excess dispersion in the EACD(1,0) model is defined as

1
V [Di ]/E[Di ]2 =
1 − 2α2
so that the ratio of the standard deviation to the mean duration is above one if α > 0.

4 Test Implementation
We will first discuss the specific implementation of the hit sequence tests suggested above. Later,
we will simulate observations from a realistic portfolio return process and calculate risk measures
from the popular Historical Simulation risk model, which in turn provide us with hit sequences
for testing.

4.1 Implementing the Markov Tests

The log-likelihood function for a sample of T i.i.d. observations from a Bernoulli variable, It ,
with known probability p is written as

ln L (I, p) = pT1 (1 − p)T −T1

where T1 is the number of ones in the sample. The log-likelihood function for an i.i.d. Bernoulli
with unknown probability parameter, π 1 , to be estimated is

ln L (I, π 1 ) = π T1 1 (1 − π 1 )T −T1 .

6
The ML estimate of π 1 is
π̂ 1 = T1 /T

and we can thus write a likelihood ratio test of unconditional coverage as

LRuc = −2 (ln L (I, π̂1 ) − ln L (I, p)) .

For the independence test, the likelihood under the alternative hypothesis is

ln L (I, π 01 , π 11 ) = (1 − π01 )T0 −T01 πT0101 (1 − π 11 )T1 −T11 π T1111

where Tij denotes the number of observations with a j following an i. The ML estimates are

π̂ 01 = T01 /T0
π̂ 11 = T11 /T1

and the independence test statistic is

LRind = −2 (ln L (I, π̂ 01 , π̂ 11 ) − ln L (I, π̂ 1 )) .

Finally the test of conditional coverage is written as

LRcc = −2 (ln L (I, π̂ 01 , π̂ 11 ) − ln L (I, p)) .

We note that all the tests are carried out conditioning on the first observation. The tests are
asymptotically distributed as χ2 with degree of freedom one for the uc and ind tests and two for
the cc test. But we will instead rely on finite sample p-values below.
Finally, as a practical matter, if the sample at hand has T11 = 0, which can easily happen in
small samples and with small coverage rates, then we calculate the first-order Markov likelihood
as
ln L (I, π01 , π 11 ) = (1 − π 01 )T0 −T01 π T0101

and carry out the tests as above.

4.2 Implementing the Weibull and EACD Tests

In order to implement our tests based on the duration between violations we first need to trans-
form the hit sequence into a duration series Di . While doing this transformation we also create
the series Ci to indicate if a duration is censored (Ci = 1) or not (Ci = 0). Except for the first
and last duration the procedure is straightforward, we just count the number of days between
each violation and set Ci = 0. For the first observation if the hit sequence starts with 0 then

7
D1 is the number of days until we get the first hit. Accordingly C1 = 1 because the observed
duration is left-censored. If instead the hit sequence starts with a 1 then D1 is simply the number
of days until the second hit and C1 = 0.
The procedure is similar for the last duration. If the last observation of the hit sequence is
0 then the last duration, DN(T ) , is the number of days after the last 1 in the hit sequence and
CN(T ) = 1 because the spell is right-censored. In the same manner if the last observation of the
hit sequence is a 1 then DN(T ) = tN(T ) − tN(T )−1 and CN (T ) = 0.
The contribution to the likelihood of an uncensored observation is its corresponding p.d.f.
For a censored observation, we merely know that the process lasted at least D1 or DN (T ) so the
contribution to the likelihood is not the p.d.f. but its survival function S(Di ) = 1 − F (Di ).
Combining the censored and uncensored observations, the log-likelihood is
N(T )−1
X
L(D; Θ) = C1 ln S(D1 ) + (1 − C1 ) ln f (D1 ) + ln(f (Di ))
i=2
+CN(T ) ln S(DN (T ) ) + (1 − CN(T ) ) ln f (DN(T ) ).

Once the durations are computed and the truncations taken care of, then the likelihood
ratio tests can be calculated in a straightforward fashion. The only added complication is that
the ML estimates are no longer available in closed form, they must be found using numerical
optimization.

4.3 Finite Sample Inference

While the large-sample distributions of the likelihood ratio tests we have suggested above are
well-known,4 they may not lead to reliable inference in realistic risk management settings. The
nominal sample sizes can be reasonably large, say two to four years of daily data, but the scarcity
of violations of for example the 1% V aR renders the effective sample size small. In this section,
we therefore present the technique of Monte Carlo tests [see Dufour (2000)].
For the case of a continuous test statistic, the procedure is the following. We first generate
N independent realizations of the test statistic, LRi , i = 1, . . . , N . We denote by LR0 the test
computed with the original sample. Under the hypothesis that the risk model is correct we
know that the hit sequence is i.i.d. Bernoulli with the mean equal to the coverage rate in our
application. We thus benefit from the advantage of not having nuisance parameters under the
null hypothesis.
4
Testing α = 0 in the EACD(1,0) model presents a potential difficulty asymptotically in that it is on the
boundary of the parameter space.

8
 We next rank LRi , i = 0, . . . , N in non-decreasing order and obtain the Monte Carlo p-value
p̂N (LR0 ) where
N ĜN (LR0 ) + 1
p̂N (LR0 ) =
N +1
with
1 X
N
ĜN (LR0 ) = 1 (LRi > LR0 )
N i=1
where 1 (∗) takes on the value 1 if ∗ is true and the value 0 otherwise.
When working with binary sequences the test values can only take a countable number of
distinct values. Therefore, we need a rule to break ties between the test value obtained from
the sample and those obtained from Monte Carlo simulation under the null hypothesis. The
tie-breaking procedure is as follows: For each test statistic, LRi , i = 0, . . . , N , we draw an
independent realization of a Uniform distribution on the [0; 1] interval. Denote these draws by
Ui , i = 0, . . . , N . The Monte-Carlo p-value is now given by

N G̃N (LR0 ) + 1
p̃N (LR0 ) =
N +1
with
1 X 1 X
N N
G̃N (LR0 ) = 1 − 1 (LRi < LR0 ) + 1 (LRi = LR0 ) 1 (Ui ≥ U0 ) .
N i=1 N i=1

5 Backtesting V aRs from Historical Simulation

We now assess the power of the proposed duration tests in the context of a Monte Carlo study.
Consider a portfolio where the returns are drawn from a GARCH(1,1)-t(d) model with leverage,
that is
p
Rt+1 = σ t+1 ((ν − 2) /ν)zt+1 , with
³p ´2
σ 2t+1 = ω + ασ 2t ((ν − 2) /ν)zt − θ + βσ 2t

where the innovation zt+1 ’s are drawn independently from a Student’s t (ν) distribution. Notice
that the innovations have been rescaled to ensure that the conditional variance of return will be
σ 2t+1 .

9
 In the simulations below we choose the following parameterization

α = 0.1
θ = 0.5
β = 0.85
ω = 3.9683e − 6
ν = 8

where ω is set to target an annual standard deviation of 0.20. The parameters imply a daily
volatility persistence of 0.975, a mean of zero, a conditional skewness of zero, and a conditional
(excess) kurtosis of 1.5. This particular DGP is constructed to form a realistic representation of
an equity portfolio return distribution.
The risk measurement method under study is the popular Historical Simulation (HS) tech-
nique. It takes the Value at Risk on a certain day to be simply the unconditional quantile of the
past Te daily observations. Specifically
p
V aRt+1 = −P ercentile({Rτ }tτ =t−Te +1 , 100p).

From the return sample and the above V aR, we are implicitly assuming that $1 is invested
each day. Equivalently, the V aR can be interpreted as being calculated in percent of the portfolio
value.
In practice, the sample size is often determined by practical considerations such as the amount
of effort involved in valuing the current portfolio holdings using past prices on the underlying
securities. For the purposes of this Monte Carlo experiment, we set Te = 250 or Te = 500
corresponding to roughly one or two years of trading days.
The V aR coverage rate, p, is typically chosen in practice to be either 1% or 5%, and below
we assess the power to reject the HS model using either of those rates. Figure 2 shows a return
sample path from the above GARCH-t(d) process along with the 1% and 5% V aRs from the HS
model (with Te = 500). Notice the peculiar step-shaped V aRs resulting from the HS method.
Notice also the infrequent changes in the 1% V aR.
The V aR exceedences from the return sample path and the 1% V aR are shown in Figure 3
reported in daily standard deviations of returns. The simulated data in Figure 3 can thus be
compared with the real-life data in Figure 1, which was taken from Berkowitz and O’Brien (2002).
The clustering and the magnitude of the exceedences are quite similar across the two plots. Note
that we have simulated 1,000 observations in Figure 3, while Figure 1 contains between 550 and
750 observations per bank. Figure 3 contains more violations than Figure 1 because of these
differences in the sample size and because the banks in Figure 1 tend to report V aRs which on
average lead to fewer than p violations.

10
 Before assessing the finite sample power results we simulate one very long realization (5
million observations) of the GARCH return process and calculate 1% and 5% V aRs from His-
torical Simulation with a rolling set of 500 in-sample returns. The zero-one hit sequence is then
calculated from the ex-post daily returns and the ex-ante V aRs.
Figure 4 plots the hazard functions of the duration between violations in the long simulation
of GARCH data and Historical Simulation V aRs. The hazard from the 1% V aR is shown in
the top panel, and the 5% V aR in the bottom panel. The hazard functions are estimated
nonparametrically via the Kaplan-Meier product-limit estimator of the survival function, which
is described in Kiefer (1988). These hazards are estimated over intervals of 15 days so if there is
a probability p of getting a hit at each day then the probability that a given duration will last
less than 15 days is
X
15 X
15
Pr(D = i) = (1 − p)i−1 p
i=1 i=1
= 1 − (1 − p)15 .

For p equal to 1% and 5% we get a constant hazard of 0.14 and 0.54 respectively over a 15-day
interval. We see in Figure 4 that the estimated hazard is at first bigger and then lower than
what we would get with a constant probability of getting a hit. Notice the distinctly downward
sloping hazard functions, which correspond to positive duration dependence. Finally, Figure 5
shows the simple histograms of durations between the violations. The top panel again shows the
1% V aR and the bottom panel shows the 5% V aR.
Data and other resource constraints often force risk managers to backtest their models on a
relatively limited backtesting samples. We therefore conduct our power experiment with samples
sizes from 500 to 1500 days in increments of 250 days. Thus our backtesting samples correspond
to approximately two through six years.
Below we simulate GARCH returns, calculate HS V aRs and the various test statistics over
1,000 Monte Carlo replications. The power of the tests are then simply calculated as the number
of simulations, divided by 1000, in which the Monte Carlo p-value is smaller than the chosen
level. The rejection frequencies are calculated at the 1%, 5% and 10% significance levels. In
order to compute p-values we simulate N = 9999 hit sequence samples under the null hypothesis
that the sequences are distributed i.i.d. Bernoulli(p).
In order to make sure that we can calculate the test statistics, we do not use Monte Carlo
samples with zero or one V aR violations.5 This of course constitutes a nontrivial sample selection
rule for the smallest sample sizes and the 1% V aR coverage rate. As it is done for all the tests
5
The likelihood of the Weibull distribution can be unbounded when we have only one uncensored observation.
When it happens we discard the sample. We get an unbounded likelihood for less than 3% of the draws when the

11
considered, the results are still comparable across tests. It also appears to be realistic that a risk
management team would not start backtesting unless at least a couple of violations had occurred.
The rejection frequencies below reflect this sample selection which is particularly important for
the low (e.g. 1%) V aR coverage rates and in the smallest samples (500 observations).

5.1 Results
The results of the Monte Carlo simulations are presented in Tables 1 and 2. We report the
empirical rejection frequencies (power) for the Markov, Weibull and EACD independence tests
for various significance test levels, V aR coverage rates, and backtesting sample sizes. Table 1
reports power for a Historical Simulation Risk model with Te = 500 rolling estimation sample
observations and Table 2 for Te = 250 rolling estimation sample observations.6
The results are quite striking. The main result is that the Weibull test is virtually almost
more powerful than the Markov and EACD tests in rejecting the HS risk models. This result
holds across inference sample sizes, VaR coverage rates and significance levels chosen. The only
two exceptions occur in Table 1 for a significance level of 1%, a coverage rate of 5% and a sample
of 500 where the EACD is better and in Table 1 for a significance level of 1%, a coverage rate of
1% and a sample of 500 where the Markov test is slightly better.
The differences in power are sometimes very large. For example in Table 1 using a 1%
significance level, the 5% VaR in a sample of 1,250 observations has a Weibull rejection frequency
of 65.2% and a Markov rejection frequency of only 29.8%. The Weibull test clearly appears to
pick up dependence in the hit violations which is ignored by the Markov test.
The performance of the EACD test on the other hand is quite sporadic. It appears to do
quite well at smaller sample sizes but relatively poorly at larger sample sizes. We suspect that
the nonlinear estimate of the α parameter is poorly behaved.
Note that the rejection frequencies for a given test are not always increasing in inference
sample size. This is due to the sample selection procedure in which we discard samples with less
than two violations. This sample selection is going to increase power ceteris paribus, and it is
going to have the biggest effect in cells corresponding to the fewest average number of violations.
These are of course the smallest sample sizes and the smallest coverage rate.
Comparing rejection frequencies across coverage rates in Table 1 we also note that in small
samples the power is sometimes higher for the 1% VaR coverage rate than for the corresponding
coverage rate is 1% and the sample size is 500, and the probability is smaller than 0.5% for higher coverage rates
and sample sizes.
6
We focus solely on the independence tests here because the historical simulation risk models under study are
correctly specified unconditionally.

12
5% VaR. This may appear to be surprising as the hit sequences from the 5% VaRs contain many
more violations which is the source of power. But the sample selection procedure will again have
the largest effect for the 1% coverage rate and for the smallest samples. The selected samples for
1% coverage will thus tend to display more dependence on average than those selected for 5%
coverage rate.
Comparing numbers across Tables 1 and 2, we note that for a coverage rate of 1% the HS VaR
with Te = 500 rolling sample observations always has a higher rejection frequency than the HS
VaR with Te = 250 rolling sample observations. This result is interesting because practitioners
often work very hard to expand their data bases enabling them to increase their rolling estimation
sample period. Our results indicate that such efforts may be futile. When the return volatility
process is very persistent, it is better to use a relatively short rolling estimation sample period.

6 Possible Extensions
6.1 The Monte Carlo Study
Encouraged by the results in Tables 1-2, we now briefly outline some possible extensions to the
Monte Carlo study:

• We only investigated one particular parameterization of the GARCH process above. It

may be interesting to calculate the power of the test for processes with different volatility
persistence, different degrees of conditional kurtosis and different leverage effects.

• One could also consider more elaborate data generating processes. Engle and Lee (1999)
consider a component GARCH model which delivers long-memory like patterns in volatility.
Hansen (1994) considers GARCH-t(vt ) models where the degrees of freedom, ν t , varies over
time in an autoregressive fashion.

• Structural breaks in the underlying return models, such as those investigated by Andreou
and Ghysels (2002), may be of interest as well.

• Hamilton and Jorda (2002) have recently introduced a class of dynamic hazard models.
Exploring these for the purpose of backtesting could be interesting.

Finally, before even venturing into the backtesting of actual risk models it may be useful
to conduct a more basic Monte Carlo analysis drawing violation sequences and duration data
directly. Specifically, if the violation sequence is generated by a first-order Markov process, what
is then the power of the different tests? Conversely, if the violation sequence is constructed from
simulated duration data with dependence, then what would the power of the different tests be?

13
6.2 Backtesting Tail Density Forecasts
The choice of Value-at-Risk as a portfolio risk measure can be criticized on several fronts. Most
importantly, the quantile nature of the V aR implies that the shape of the return distribution
to the left of the left is ignored. Particularly in portfolio’s with highly nonlinear distributions,
such as those including options, this shortcoming can be crucial. Theoreticians have criticized
the V aR measure both from a utility-theoretic perspective (Artzner et al, 1999) and from a
dynamic trading perspective (Basak and Shapiro, 2000). Although some of these criticisms have
recently been challenged (Cuoco, He, and Issaenko, 2001), it is safe to say that risk managers
ought to be interested in knowing the entire distribution of returns, and in particular the left
tail. Backtesting distributions rather than V aRs then becomes important.
Consider the standard density forecast evaluation approach7 of calculating the uniform trans-
form variable
Ut = Ft (Rt )

where Ft (∗) is the a priori density forecast for time t. The null hypothesis that the density
forecast is optimal corresponds to

Ut ∼ i.i.d. Unif orm(0, 1)

Berkowitz (2001) argues that the bounded support of the uniform variable renders standard
inference difficult. One is forced to rely on nonparametric tests which have notoriously poor
small sample properties. He suggests a simple transformation using the inverse normal c.d.f.

Zt = Φ−1 (Ut )

after which the hypothesis

Zt ∼ i.i.d. Normal(0, 1)

can easily be tested.

Berkowitz further argues that confining attention to the left tail of the distribution has par-
ticular merit in the backtesting of risk models where the left tail contains the largest losses, which
are most likely to impose bankruptcy risk. He defines the censored variable
(
∗ Zt , if Rt < V aRt
Zt =
Φ−1 (V aRt ) , else

and tests the null that

Zt∗ ∼ Censored Normal(0, 1, V aRt )
7
See for example Diebold, Gunther and Tay (1998).

14
 We note first that Berkowitz (2001) only tests the unconditional distribution of Zt∗ . The
information in the potential clustering of the V aR exceedences is ignored.
Second, note that the censored variable complication is not needed. If we want to test that
the transforms of the p100 largest losses are themselves uniform, then we can simply multiply the
subset of the uniform by 1/p, apply the transformation and test for standard normality again.8
That is (
∗∗ Ut /p, if Rt < V aRt
Ui =
Else not defined
We then have that
Zi∗∗ = Φ−1 (Ui∗∗ ) ∼ i.i.d. Normal(0, 1)

Note that due to the censoring there is no notion of time in the sequence Zi∗∗ . We might
want to make a joint analysis of both Zi∗∗ and the duration between violations Di . To do this
we would like to write a joint density for these two processes under the alternative. We know
that under the null hypothesis that the risk model is correctly specified the Zi∗∗ should be i.i.d.
N(0, 1), Di should be i.i.d. exponential with mean 1/p, and the processes should be independent.
The question is how to write a joint density for these two processes as the alternative hypothesis
knowing that, for example, the marginal p.d.f. of Di is a Weibull and some other p.d.f. for Zi∗∗ ?
Copulas provide a useful tool for doing so.
A (bivariate) copula is a function C from [0; 1] × [0; 1] to [0; 1] with the following properties:

1. For every u, v in [0; 1],

C(u, 0) = 0 = C(0, v)

and
C(u, 1) = u and C(1, v) = v;

2. For every u1 , u2 , v1 , v2 in [0; 1] such that u1 ≤ u2 and v1 ≤ v2 ,

C(u2 , v2 ) − C(u2 , v1 ) − C(u1 , v2 ) + C(u1 , v1 ) ≥ 0

In order to explain how copulas can be used we apply Sklar’s theorem (Nelsen, 1998), which
states: Let H be a joint distribution function with margins F and G. Then there exists a copula
C such that for all x, y in R,
H(x, y) = C(F (x), G(y)).

If F and G are continuous then C is unique. Conversely, if C is a copula and F and G are
distribution functions then H is a joint distribution function with marginal densities F and G.
8
We are grateful to Nour Meddahi for pointing this out.

15
 So if we have two densities under the alternative (e.g. f (Di ) and g(Zi∗∗ )) then we can easily
construct a joint density by applying a copula. Suppose the considered bivariate copula C(u, v; θ)
is a function of a unique parameter θ and that we have C(u, v; θ0 ) = uv and C(u, v; θ) 6= uv for
θ 6= θ0 . This gives us a basis for a test because C(F (x), G(y); θ0 ) = F (x)G(y) means that x and
y are independent.
An example of such a copula is the Ali-Mikhail-Haq family of copulas where
uv
C(u, v; θ) = ; θ ∈ [−1, 1]
1 − θ(1 − u)(1 − v)

and we have C(u, v; θ) = uv if θ = 0. A possible alternative hypothesis could be that Di is i.i.d.

Weibull(a, b), Zi∗∗ is i.i.d. N(µ, σ 2 ) and C(u, v; θ) is from the Ali-Mikhail-Haq family of copulas.
We could then test

H0 : a = p, b = 1, µ = 0, σ = 1, θ = 0
H1 : at least one of these equalities does not hold

in a likelihood ratio framework similar to the one considered for the V aR tests above. We plan
to the pursue the implementation of such tests in future work.

7 Summary
We have presented a new set of procedures for backtesting risk models. The chief insight is
that if the V aR model is correctly specified for coverage rate, p, then the conditional expected
duration between violations should be a constant 1/p days. We suggest various ways of testing
this null hypothesis and we conduct a Monte Carlo analysis which compares the new tests to those
currently available. Our results show that in many of the situations we consider, the duration
based tests have much better power properties than the previously suggested tests. The size
of the tests is easily controlled through finite sample p-values, which we calculate using Monte
Carlo simulation.
The immediate potential extensions to our Monte Carlo results are many. We could consider
alternative data generating processes for returns and alternative risk models. Allowing for re-
alistic nonstationarities such as structural breaks in the return process could be interesting as
well.
The majority of financial institutions use V aR as a risk measure, and many calculate VaR
using the so-called Historical Simulation approach. While the main focus of our paper has thus
been backtesting V aRs from Historical Simulation, we also suggest extensions to density and
density tail backtesting.

16
References
[1] Andreou and Ghysels (2002), Quality Control for Value at Risk: Monitoring Disruptions in
the Distribution of Risk Exposure, Manuscript, University of North Carolina.

[2] Artzner, P., F. Delbaen, J.-M. Eber and D. Heath (1999), Coherent Measures of Risk,
Mathematical Finance, 9, 203-228.

[3] Barone-Adesi, G., K. Giannopoulos and L. Vosper (2000), Backtesting Derivative Portfolios
with FHS, Manuscript, USI and City Business School.

[4] Basak, S. and A. Shapiro (2000), Value at Risk Based Risk Management: Optimal Policies
and Asset Prices, Review of Financial Studies, 14, 371-405.

[5] Basle Committee on Banking Supervision (1996), Amendment to the Capital Accord to
Incorporate Market Risks. Basle.

[6] Beder, T. (1995), VaR: Seductive but Dangerous, Financial Analysts Journal, September-
October, 12-24.

[7] Berkowitz, J. (2001), Testing Density Forecasts, Applications to Risk Management Journal
of Business and Economic Statistics, 19, 465-474.

[8] Berkowitz, J. and J. O’Brien (2002), How Accurate are the Value-at-Risk Models at Com-
mercial Banks? Journal of Finance, 57, 1093-1112.

[9] Christoffersen, P. (1998), Evaluating Interval Forecasts, International Economic Review, 39,
841-862.

[10] Christoffersen, P. (2002), Elements of Financial Risk Management, Academic Press, Forth-
coming.

[11] Christoffersen, P., J. Hahn and A. Inoue (2001), Testing and Comparing Value-at-Risk
Measures Journal of Empirical Finance, 8, 325-342.

[12] Cuoco, D., H. He, and S. Issaenko (2001), Optimal Dynamic Trading Strategies with Risk
Limits, Manuscript, Yale University.

[13] Diebold, F.X., T. Gunther, and A. Tay (1998), Evaluating Density Forecasts, with Appli-
cations to Financial Risk Management, International Economic Review, 39, 863-883 .

17
[14] Dufour, J.-M. (2000), Monte Carlo Tests with Nuisance Parameters : A General Approach
to Finite-Sample Inference and Nonstandard Asymptotics in Econometrics, Manuscript,
Université de Montréal.

[15] Engle, R. and G.J. Lee (1999), A Permanent and Transitory Component Model of Stock
Return Volatility, in ed. R. Engle and H. White Cointegration, Causality, and Forecasting:
A Festschrift in Honor of Clive W.J. Granger, Oxford University Press, 475-497.

[16] Engle, R. and J. Russel (1998), Autoregressive Conditional Duration: A New Model for
Irregularly Spaced Transaction Data, Econometrica, 66, 1127-1162.

[17] Gourieroux, C. (2000) Econometrics of Qualitative Dependent Variables. Translated by Paul

B. Klassen. Cambridge University Press.

[18] Hamilton, J. and O. Jorda (2002), A Model for the Federal Funds Rate Target, Journal of
Political Economy. Forthcoming.

[19] Hansen, B. (1994), Autoregressive Conditional Density Estimation, International Economic

Review, 35, 705-730.

[20] Hendricks, D. (1996), Evaluation of Value-at-Risk Models Using Historical Data, Economic
Policy Review, Federal Reserve Bank of New York, April, 39-69.

[21] Kiefer, N. (1988), Economic Duration Data and Hazard Functions, Journal of Economic
Literature, 26, 646-679.

[22] Kupiec, P. (1995), Techniques for Verifying the Accuracy of Risk Measurement Models,
Journal of Derivatives, 3, 73-84.

[23] Jorion, P. (2000), Value-at-Risk: The New Benchmark for Controlling Financial Risk.
Chicago: McGraw-Hlill.

[24] Nelsen, R.(1998), An Introduction to Copulas, Lectures Notes in Statistics, 139, Springer
Verlag.

[25] Poirier, D. (1995), Intermediate Statistics and Econometrics: A Comparative Approach.

Cambridge, MA: MIT Press.

[26] Pritsker, M. (1997), Evaluating Value at Risk Methodologies: Accuracy versus Computa-
tional Time, Journal of Financial Services Research, 201-241.

[27] Pritsker, M. (2001), The Hidden Dangers of Historical Simulation, Manuscript, Federal
Reserve Board.

18
 Figure 1
Value-at-Risk Exceedences
From Six Major Commercial Banks
Berkowitz and O’Brien (2002)

19
 Figure 2
GARCH-t(d) Simulated Portfolio Returns with
1% and 5% Value-at-Risk from Historical Simulation with Te = 500

0.08

0.06

0.04

0.02

-0.02

-0.04

-0.06

-0.08

-0.1
0 100 200 300 400 500 600 700 800 900 1000

20
 Figure 3
GARCH-t(d) Simulated Portfolio Returns with
Exeedences of 1% V aRs from Historical Simulation with Te = 500
Reported in Standard Deviations of Returns

-0.5

-1

-1.5

-2

-2.5

-3

-3.5
0 100 200 300 400 500 600 700 800 900 1000

21
 Figure 4
Hazard Functions of Duration between V aR Violations
GARCH-t(d) Portfolio Returns
Historical Simulation Risk Model with Te = 500

1% Value-at-Risk
1

0.8

0.6

0.4

0.2

0
50 100 150 200 250

5% Value-at-Risk
1

0.8

0.6

0.4

0.2

0
50 100 150 200 250

22
 Figure 5
Histograms of Duration between V aR Violations
GARCH-t(d) Portfolio Returns
Historical Simulation Risk Model with Te = 500

1% Value-at-Risk
1

0.8

0.6

0.4

0.2

0
25 75 125 175 225 275 325 375 425 475 525 575 625 675 725

5% Value-at-Risk
1

0.8

0.6

0.4

0.2

0
25 75 125 175 225 275 325 375 425 475 525 575 625 675 725

23
 Table 1: Empirical Power in Independence Tests: Historical Simulation with Te = 500

Significance Level: 1% Significance Level: 5% Significance Level: 10%

Coverage Rate: 1% Coverage Rate: 1% Coverage Rate: 1%
Test: Markov Weibull EACD Test: Markov Weibull EACD Test: Markov Weibull EACD
Sample size Sample size Sample size
500 0.1190 0.1790 0.1530 500 0.3320 0.3520 0.2510 500 0.4210 0.4690 0.3100
750 0.1450 0.2510 0.1840 750 0.2940 0.4850 0.2560 750 0.4620 0.5840 0.3270
1000 0.1950 0.3800 0.1240 1000 0.3320 0.5900 0.2300 1000 0.4960 0.6730 0.2770
1250 0.2480 0.4840 0.1600 1250 0.3750 0.6750 0.2590 1250 0.5090 0.7550 0.3220
1500 0.2930 0.6030 0.1300 1500 0.4020 0.7550 0.2150 1500 0.5310 0.8200 0.2600

Coverage Rate: 5% Coverage Rate: 5% Coverage Rate: 5%

Test: Markov Weibull EACD Test: Markov Weibull EACD Test: Markov Weibull EACD
Sample size Sample size Sample size
500 0.2120 0.2770 0.3290 500 0.3010 0.4560 0.4320 500 0.3600 0.5390 0.4880
750 0.2720 0.4610 0.4030 750 0.3690 0.6410 0.5170 750 0.4420 0.7390 0.5940
1000 0.3090 0.6070 0.4120 1000 0.4090 0.7670 0.5630 1000 0.4920 0.8280 0.6280
1250 0.3970 0.6760 0.5220 1250 0.5530 0.8370 0.6380 1250 0.6720 0.8920 0.6970
1500 0.4190 0.7650 0.4840 1500 0.6360 0.8970 0.6180 1500 0.7220 0.9330 0.6800
 Table 2: Empirical Power in Independence Tests: Historical Simulation with Te = 250

Significance Level: 1% Significance Level: 5% Significance Level: 10%

Coverage Rate: 1% Coverage Rate: 1% Coverage Rate: 1%
Test: Markov Weibull EACD Test: Markov Weibull EACD Test: Markov Weibull EACD
Sample size Sample size Sample size
500 0.0990 0.1040 0.0720 500 0.2460 0.2560 0.1540 500 0.2830 0.3530 0.2030
750 0.0940 0.0890 0.0560 750 0.2340 0.2880 0.1100 750 0.3050 0.4100 0.1650
1000 0.1110 0.1690 0.0420 1000 0.2720 0.3480 0.1100 1000 0.3750 0.4800 0.1430
1250 0.1390 0.2240 0.0240 1250 0.2990 0.4620 0.0700 1250 0.4080 0.5630 0.1120
1500 0.1880 0.3350 0.0180 1500 0.3200 0.5360 0.0590 1500 0.4610 0.6370 0.0960

Coverage Rate: 5% Coverage Rate: 5% Coverage Rate: 5%

Test: Markov Weibull EACD Test: Markov Weibull EACD Test: Markov Weibull EACD
Sample size Sample size Sample size
500 0.1970 0.3030 0.2990 500 0.2830 0.4660 0.4310 500 0.3480 0.5520 0.4780
750 0.2540 0.4230 0.3510 750 0.3720 0.6360 0.4790 750 0.4100 0.7300 0.5380
1000 0.3060 0.5670 0.3470 1000 0.4150 0.7420 0.4750 1000 0.5070 0.8170 0.5340
1250 0.2980 0.6520 0.3570 1250 0.4890 0.8110 0.4880 1250 0.6070 0.8680 0.5380
1500 0.3700 0.7300 0.3830 1500 0.6020 0.8770 0.5280 1500 0.7120 0.9150 0.6130