Professional Documents
Culture Documents
2012 - Trade Sizing Techniques For Drawdown and Tail Risk Control
2012 - Trade Sizing Techniques For Drawdown and Tail Risk Control
Issam S. S TRUB ∗
Abstract
This article introduces three algorithms for trade sizing with the objective of controlling tail risk or
maximum drawdown when applied to a trading strategy. The first algorithm relies on historical volatility
estimates while the second uses tail risk estimates obtained by applying Extreme Value Theory (EVT) to
estimate Conditional Value at Risk (CVaR); the third algorithm also uses Extreme Value Theory applied
to the drawdown distribution to compute the Conditional Drawdown at Risk (CDaR). These algorithms
are applied to 10 years of daily returns from a trend following strategy trading the EURUSD and NZD-
MXN currency pairs. In each case, the performance of the algorithms is analysed in detail and compared
to the original strategy. The ability of these algorithms in terms of tail risk and drawdown control is eval-
uated. The techniques presented in the article are readily applicable by investment managers to compute
adequate trade size while maintaining a constant level of tail risk or limiting maximum drawdown to a
chosen value.
1 Introduction
Money management, also called position or trade sizing, consists in selecting an appropriate leverage level
to be applied to a given strategy, where the leverage level is defined as the ratio of the position market value
with respect to the total assets under management (AUM) or account size. At a glance, it seems quite obvi-
ous that the ability to dynamically adjust position size can result in increased profits; indeed, an investment
manager able to reduce (respectively increase) leverage before periods of underperformance (respectively
outperformance) would obtain higher returns when compared to using a constant leverage. Practical evi-
dence confirms that most traders tend to dynamically adjust their leverage level , usually relying on heuristic
rules and being subject to behavioural biases, as demonstrated in L OCKE and M ANN (2003), T HALER and
J OHNSON (1990). However, while a significant literature exists dealing with trading strategies profitabil-
ity, there are fewer articles on money management and the limited literature on this topic often presents
techniques which are not always applicable in practice by traders. Additionally, these money management
techniques rarely make use of modern econometrics and risk management tools such as Extreme Value
Theory or time series analysis.
Historically, money management has been applied to gambling as well as trading, typically with the
aim of maximising growth rate. Indeed, in K ELLY (1956), information theory and expected utility function
theory (introduced in B ERNOULLI (1738, 1954) in relation to the St. Petersburg paradox) were combined
∗
Research Scientist, strub@cal.berkeley.edu
to obtain an optimal gambling strategy, the Kelly criterion: maximise the expected value of the logarithm
of the gambler’s wealth at each bet to achieve an asymptotically optimal growth rate; such a strategy also
minimises the expected time to reach a given wealth as demonstrated in B REIMAN (1961) in the case where
stock returns are assumed to be independent, identically distributed (i.i.d.). These results were applied to
investment management in L ATAN É (1959), who advised investors to maximise the geometric mean of their
portfolios. Later, the optimality of maximising expected log return was extended with no restrictions on the
distribution of the market process in A LGOET and C OVER (1988) and, in B ROWNE and W HITT (1996), the
Kelly criterion was generalised to the case in which the underlying stochastic process is a simple random
walk in a random environment; O SORIO (2009) devised an analog to the Kelly criterion for fat tail returns
modelled by a Student t-distribution and a log prospect rather than utility function.
In H AKANSSON (1970), closed form optimal consumption, investment and borrowing strategies were
obtained for a class of utility functions corresponding to an investor looking to maximise the expected utility
from consumption over time given an initial capital position and a known deterministic non-capital income
stream. ROLL (1973) studied the implications of growth optimum portfolios in terms of observed stock
returns for investors selecting such a portfolio. T HORP (1971) applied the Kelly criterion of maximising
logarithmic utility to portfolio choice and compared it to mean variance portfolio theory, concluding that the
Kelly criterion does not always yield mean variance efficient portfolios.
S AMUELSON (1971, 1979) showed that while maximising the geometric mean utility at each stage may
be asymptotically optimal, this does not imply that such a strategy is optimal in finite time; he also high-
lighted the risk involved in using the Kelly criterion, namely that of excessive leverage leading to significant
drawdowns. Later on, more work on the properties of the Kelly criterion and its application to finance was
published in ROTANDO and T HORP (1992); M AC L EAN et al. (2004, 2010, 2011a,b); T HORP (2006). The
concept of optimal f, which is an extension of the Kelly criterion was developed in V INCE (2007, 2009)
and the differences between the two approaches were detailed in V INCE (2011). Additionally, a number of
books on money management aimed at practitioners have analysed and backtested the Kelly criterion, the
optimal f and other approaches; see for example G EHM (1983); BALSARA (1992); G EHM (1995); J ONES
(1999); S TRIDSMAN (2003); M C D OWELL (2008). Finally, the optimal f technique was applied to futures
trading and compared to more naive approaches in A NDERSON and FAFF (2004); L AJBCYGIER and L IM
(2007), each time with the conclusion that it resulted in leverage levels that would be unacceptable to most
investors; this leads to heuristic approaches such as using a fraction of the Kelly ratio (such as half Kelly
ratio).
The common feature of traditional money management techniques such as the Kelly criterion or the
optimal f is their focus on maximising wealth growth, whereas in practice, both individual traders and fund
managers are mostly concerned with maintaining a stable risk level through time and keeping their maxi-
mum drawdown below a chosen threshold; otherwise, they will most likely suffer significant redemptions
from investors or discontinue trading their strategy altogether. As such, maximising the rate of return is
usually secondary to controlling risk. However, there is very little available on this topic in the existing lit-
erature whether originating from academics or practitioners. Note that the practical shortcomings of utility
maximisation had already been noted in ROY (1952) before the introduction of the Kelly criterion as the
author explained that for the average investor, the first objective is to limit the risk of a disaster occurring:
“In calling in a utility function to our aid, an appearance of generality is achieved at the cost of a loss of
practical significance and applicability in our results. A man who seeks advice about his actions will not be
grateful for the suggestion that he maximises expected utility.”
The techniques presented in this article were born out of the need for position sizing rules that could be
computed and applied in practice and would result in consistent risk levels when implemented through varied
market conditions and changing trading strategy performance. Algorithms designed to control return tail risk
are presented first, while drawdown control is tackled at a later stage. These techniques are then applied to
daily returns resulting from implementing a simple technical trading rule on the EURUSD and NZDMXN
currency pairs; a detailed analysis and comparison of the performance of each money management technique
concludes the article.
15
14 Normal Distribution
13
12
Number of Observations
11
10
9
8 95% VaR
7
6
Generalised Pareto Distribution
5
4
3
2
1
0
−50 −45 −40 −35 −30 −25 −20 −15 −7.8−5 0
Daily Return (%)
Figure 1: Top: Comparison of Generalised Pareto and normal distribution. Note that the Generalised Pareto
Distribution models the left tail of the daily returns much more accurately than the normal distribution.
Bottom: 95 % CVaR for each distribution. The CVaR is represented by the shaded area under the green
(GPD) or red (normal distribution) curve. In the present case, it is apparent that the CVaR computed using
a normal distribution underestimates the downside risk when compared to a GPD.
4
2.2 Volatility based Position Sizing 2 TAIL RISK CONTROL
where T is the length of the estimation window, λ the decay factor and r the mean return over the estimation
window.
Once the volatility has been computed it can be converted into a VaR number using Equation (3) and the
leverage or position size is adjusted through the formula:
Target VaR
Leverage Adjustment = (5)
Current VaR
This process is typically implemented with a chosen frequency (daily, weekly, monthly) depending on
the average holding period of the trading strategy.
where β > 0, and the support of Gξ,β is y > 0 when ξ > 0 and 0 6 y 6 − βξ when ξ < 0.
The shape and scale parameters ξ and β can be estimated using Maximum Likelihood Estimation (MLE)
by fitting a GPD distribution to the tail of the return distribution after a given threshold u. Once this is done,
the CVaR can be computed:
VaRα + β − ξu
CVaRα = (7)
1−ξ
where the VaR for a GPD can be estimated by:
(( )−ξ )
β αN
VaRα = u + −1 (8)
ξ Nu
5
2.3 Extreme Value Theory based Position Sizing 2 TAIL RISK CONTROL
with N the total number of observations and Nu the number of observations exceeding the threshold u.
Note that the preceding results requires that observations be independent and identically distributed,
which is often not the case for daily returns as they present some level of autocorrelation. Therefore, we
start by filtering the daily returns and then apply Extreme Value Theory to the standardised residuals (see
M C N EIL and F REY (2000); N YSTR ÖM and S KOGLUND (2005)), with a Generalised Pareto Distribution
being fitted to the tails through Maximum Likelihood Estimation. Once this is done, we obtain the shape and
scale parameters and replace these values in Equation (7) to compute the CVaR at the required confidence
level. Extreme Value Theory has been used during the previous decade for risk management in finance with
a notable increase in the number of publications on the subject since the recent financial crisis. We refer
to BALI (2003); B EIRLANT et al. (2004); C ASCON and S HADWICK (2009); C OLES (2001); DE H AAN and
F ERREIRA (2006); E MBRECHTS (2011); G HORBEL and T RABELSI (2008, 2009); G OLDBERG et al. (2008,
2009); G UMBEL (2004); H UANG et al. (2012); L ONGIN (2000); M C N EIL and F REY (2000); N YSTR ÖM and
S KOGLUND (2005) for a sample of publications dealing with Extreme Value Theory and its applications to
financial risk modelling.
The significant improvement in tail risk modelling between the volatility/normal distribution and EVT
approaches is illustrated in Figure 1. We consider 1000 daily returns for a stock and fit both a normal
distribution and a Generalised Pareto Distribution to the left tail of the daily returns. We can see that while
both techniques yield similar VaR numbers at the 95% confidence level (in this case 7.8%), the 95% CVaR,
which can be visually identified as the area under a given distribution curve left of the 95% VaR threshold,
is significantly higher (by a factor 2.4) when computed using the Generalised Pareto Distribution than when
using volatility and a normal distribution assumption. Note that this is a pathological case which was chosen
on purpose as the difference between the two methods is readily apparent. Still, relying on volatility and
normal distribution assumptions can lead to significantly underestimating the tail risk generated by a given
strategy, a dangerous situation to be in for any investment manager.
6
2.3 Extreme Value Theory based Position Sizing 2 TAIL RISK CONTROL
time; thus recent returns have more influence than the more distant ones. Unfortunately, it is not clear how
to select the correct time constant; also, an unintended consequence is that extreme events, which by nature
occur rarely, might end up being discounted.
Second, the historical simulation method assumes that daily returns are independent and identically dis-
tributed through time, which is not particularly realistic. Indeed, it is commonly observed that the volatility
of returns evolves through time and that periods of high and low volatility do not occur at randomly spaced
intervals but rather tend to be clustered together. The filtered historical simulation (FHS) method presented
in BARONE -A DESI et al. (1999) is an attempt to combine the advantages of the historical and parametric
methods; the variance-covariance method attempts to capture conditional heteroskedasticity but assumes a
normal distribution while the historical method does not assume a specific distribution but does not capture
conditional heteroskedasticity. The FHS method relies on a model based approach for the volatility, typi-
cally using a GARCH type model, while remaining model free in terms of the distribution. In particular,
this method has the notable advantage of being able to simulate extreme losses even if they are not present
in the historical returns used for the simulation.
2
σt+1 = ω + αε2t + βσt2 , with α + β < 1 (10)
Alternately, the GARCH model can be replaced by its extension, the exponential GARCH (EGARCH)
model developed in N ELSON (1991); N ELSON and C AO (1992) to capture the asymmetry in volatility in-
duced by large positive and negative returns. Indeed, volatility usually increases more after a large drop than
after a large increase due to the leverage effect (B LACK (1976)). This model is defined by:
2
ln σt+1 = ω + α(ϕεt + γ(|εt | − E|εt |)) + β ln σt2 (11)
Once an AR(1)/GARCH(1,1) model has been fitted to the daily returns, the autocorrelation of the
squared returns is usually noticeably lower and these observations can now be used in a historical simu-
lation method. The i.i.d. property is important for bootstrapping, as it allows the sampling procedure to
safely avoid the pitfalls of sampling from a population in which successive observations are serially depen-
dent. We simulate a number of independent random trials (10,000 in this article) over a time horizon of 252
trading days; unlike Monte Carlo simulations we do not make a specific distributional assumption regarding
the standardised returns {zt } and instead use the past returns data. Given a sequence of past returns we can
compute past standardised returns from observed returns and estimated standard deviations as the quotient
7
3 DRAWDOWN CONTROL
of the residual of the AR(1) model over the standard deviation. Once the historical standardised returns are
known, we generate future returns by drawing standardised returns with replacement. Eventually, we end
up with 10,000 daily return series, each covering 252 trading days. These daily returns are aggregated to
generate a distribution of 2,520,000 daily returns to which left tail a GPD is fitted, eventually yielding the
CVaR. The high number of residuals ensures the stability of the method, as the left tail contains 126,000
returns for a 95% confidence level, which almost guarantees the convergence of the Maximum Likelihood
Estimation algorithm used to fit the GPD to the left tail of the simulated return distribution. This CVaR
number can be converted into a VaR number under normal distribution assumptions using Equation (3) and
trade size adjusted through Equation (5). One of the advantages of using Extreme Value Theory to compute
the CVaR is that the tail risk of the return distribution is measured much more accurately and less likely to
be underestimated than when relying only on the volatility based method described earlier on.
3 Drawdown Control
While the previous section outlined money management tools to control tail risk, defined as daily VaR or
CVaR at a given confidence level, the most adverse event from an investor or investment manager standpoint
is probably a significant drawdown in which a number of negative daily returns are clustered together over
a given period time. Indeed, most investors have strict drawdown limits (such as 20%) upon which they will
redeem part or the entirety of their investment in a given fund. Therefore, for a money manager, experiencing
a significant drawdown can lead to a drop in AUM which itself results in a loss of management fees; addi-
tionally, most fund managers who charge performance fees have high watermarks in place which prevent
them from collecting performance fees during a drawdown. Also, a manager trading a systematic strategy
with proprietary or investor capital is likely to unnecessarily modify or discontinue the strategy if faced with
an unacceptable drawdown; this can result in the loss of future performance as the changes may have been
unwarranted. This leads us to develop a money management technique to control the maximum drawdown
encountered by a given strategy. Earlier work on drawdown control through portfolio optimisation can be
found in C VITANIC and K ARATZAS (1995); G ROSSMAN and Z HOU (1993).
The historical maximum drawdown is a number which varies widely even for strategies presenting the
same mean and volatility and is based on the entire track record making difficult any comparison between
strategies run over different time lengths. Therefore, as noted in H ARDING et al. (2003), considering a
drawdown distribution with reference to a confidence level would be more practical. The distribution of
drawdowns over a given time period of N days can be computed by computing the maximum drawdown
for blocks of N consecutive days from the track record of a strategy. As VaR and CVaR were defined
for a daily return drawdown, the Drawdown at Risk (DaR) and Conditional Drawdown at Risk (CDaR)
at a given confidence level can be obtained from the drawdown distribution. For example, the 63 days
8
3.2 Practical Implementation 4 APPLICATIONS
DaR at the 95% confidence level will be obtained by subdividing the historical daily returns in overlapping
blocks of 63 consecutive daily returns, computing the maximum drawdown for each block thus forming the
drawdown distribution and taking the 95th percentile of this distribution. Similarly the 63 day CDaR would
be the average expected drawdown beyond the 95th percentile. The modelling of the drawdown distribution
has been considered in C HEKHLOV et al. (2003, 2005); J OHANSEN and S ORNETTE (2001); M ENDES and
B RANDI (2004); M ENDES and L EAL (2005).
Target CDaR
Leverage Adjustment = (13)
Current CDaR
4 Applications
In order to analyse the effectiveness and performance of the trade sizing algorithms defined in the previous
sections, we implement them on the daily returns generated by a systematic strategy applied to the EURUSD
and NZDMXN currency pairs.
9
4.1 Trading Strategy 4 APPLICATIONS
0.8
Sample Autocorrelation
0.6
0.4
0.2
−0.2
0 5 10 15 20
Lag
0.8
Sample Autocorrelation
0.6
0.4
0.2
−0.2
0 5 10 15 20
Lag
Figure 2: Top: The autocorrelation function of the squared daily returns for the EURUSD strategy reaches
significant values through time, thus preventing the use of unfiltered data for historical simulation. Bottom:
Autocorrelation function of the standardised residuals after filtering with an AR(1)/GARCH(1,1) model; the
autocorrelation has been almost entirely removed. 10
4.1 Trading Strategy 4 APPLICATIONS
300
250
200
150
100
0 1 2 3 4 5 6 7 8 9 10
Time (Years)
3.5
Volatility based trade sizing
EVT based trade sizing
3 CDaR based trade sizing
Leverage Adjustment Factor
2.5
1.5
0.5
0
0 1 2 3 4 5 6 7 8 9 10
Time (Years)
Figure 3: Top: Evolution of the Net Asset Value for the original EURUSD strategy and the volatility and
EVT based position sizing methods. Bottom: Evolution of the leverage adjustment factor for the volatility
and EVT based position sizing methods applied to the EURUSD strategy.
11
4.1 Trading Strategy 4 APPLICATIONS
190
Original Returns
180 Volatility based trade sizing
EVT based trade sizing
170 CDaR based trade sizing
Net Asset Value (Base 100)
160
150
140
130
120
110
100
90
0 1 2 3 4 5 6 7 8 9 10
Time (Years)
2.5
Volatility based trade sizing
EVT based trade sizing
CDaR based trade sizing
2
Leverage Adjustment Factor
1.5
0.5
0
0 1 2 3 4 5 6 7 8 9 10
Time (Years)
Figure 4: Top: Evolution of the Net Asset Value for the original NZDMXN strategy and the volatility and
EVT based position sizing methods. Bottom: Evolution of the leverage adjustment factor for the volatility
and EVT based position sizing methods applied to the NZDMXN strategy.
12
4.1 Trading Strategy 4 APPLICATIONS
200
180
160
140
120
Frequency
100
80
60
40
20
0
0 5 10 15
63 Day Drawdowns (%)
120
100
80
Frequency
60
40
20
0
0 5 10 15 20
63 Day Drawdowns (%)
Figure 5: Top: Distribution of 63 day drawdowns for the EURUSD strategy. Bottom: Distribution of 63 day
drawdowns for the NZDMXN strategy.
13
4.2 Tail Risk Control Techniques 4 APPLICATIONS
1. At the end of week N , select the previous 74 daily returns and generate the volatility using Equa-
tion (4).
3. Compute the Leverage Adjustment corresponding to a target 95% VaR of 1.5% using Equation (5).
1. At the end of week N , select the previous 252 daily returns and filter them using an AR(1)/GARCH(1,1)
model; check that the autocorrelation has been brought to a sufficiently low level for the i.i.d. assump-
tion to be valid.
2. Using the AR(1)/GARCH(1,1) model, simulate 10,000 daily returns of 252 days each, generating one
series of 2,520,000 returns after aggregation.
14
4.3 Drawdown Control Technique 4 APPLICATIONS
3. Using MLE, fit a GPD distribution to the left tail (5% worst daily returns) of the simulated return
series, yielding the shape and scale parameters.
4. Compute the 95% CVaR corresponding to the GPD parameters values and convert the CVaR into a
VaR number using Equation (3).
5. Compute the Leverage Adjustment corresponding to a target 95% VaR of 1.5% using Equation (5).
1. At the end of week N , select the previous 252 daily returns and filter them using an AR(1)/GARCH(1,1)
model; check that the autocorrelation has been brought to a sufficiently low level for the i.i.d. assump-
tion to be valid.
2. Using the AR(1)/GARCH(1,1) model, simulate 10,000 daily returns of 252 days each.
3. Decompose each series of 252 daily returns into 190 overlapping blocks of 63 consecutive days.
4. Compute the maximum drawdown for each block, and aggregated all the drawdowns into one series
of 1,900,000 drawdowns.
5. Using MLE, fit a GPD distribution to the right tail (5% largest drawdowns) of the simulated return
series, yielding the shape and scale parameters.
15
4.4 Results analysis 4 APPLICATIONS
7. Compute the Leverage Adjustment corresponding to a target 95% CDaR of 10% using Equation (13).
16
5 CONCLUSION
drawdowns for the original strategies fluctuated from 5.52% to 15.25% and from 7.21% to 17.40% respec-
tively. This demonstrates the ability to control maximum drawdown by using the CDaR based algorithm.
The evolution of the leverage factor for the CDaR based algorithm is quite smooth, making it less likely to
suffer from high transaction costs when implemented in practice. Once again, the Sharpe ratio for the CDaR
based technique is slightly higher than for the original strategies.
5 Conclusion
A number of money management techniques were presented, with the aim of controlling either tail risk or
drawdown rather than attempting to maximise return or expected utility at any cost as is the case for most
money management techniques available in the existing literature. Indeed, the main concern of investment
professionals is to remain at or below certain risk constraints set either internally or by investors; as such,
maximising expected utility is only secondary to controlling risk as a breach of these risk limits would
usually trigger significant redemptions or would lead the investment manager to stop trading the strategy
altogether.
The first two methods aim to maintain a stable level of tail risk through time, using either historical
volatility or Extreme Value Theory to measure tail risk. Both methods were applied to two sets of daily
returns generated by applying a typical trend following strategy to the EURUSD and NZDMXN currency
pairs over a 10 year period, and demonstrated the ability to target a given VaR level for the volatility based
technique or a given CVaR level for the EVT based technique. The EVT based technique, which considers
the entire left tail of the return distribution at a given confidence level, is superior to the volatility based
technique which is oblivious to the size of losses beyond the VaR threshold and therefore can result in a
higher overall tail risk than intended.
The third method focuses on drawdown control, by adjusting the leverage factor based on the Condi-
tional Drawdown at Risk level generated by the strategy. The CDaR is computed by considering overlapping
blocks of consecutive returns and calculating the maximum drawdown for each block, yielding a drawdown
distribution from which the average expected drawdown beyond a certain confidence level (CDaR) can be
obtained. Considering the drawdown distribution rather than the maximum drawdown over the entire pe-
riod results in a more stable and robust estimate of potential drawdown. The drawdown control technique
achieves its objective when applied to the two strategies as the maximum drawdown for each year remains
around or below the targeted level.
17
Return (%) Volatility (%) Max. Drawdown (%)
Strategy Orig. Vol. EVT CDaR Orig. Vol. EVT CDaR Orig. Vol. EVT CDaR
Year 1 2.28 3.19 0.56 1.94 10.96 15.26 12.02 9.75 8.08 11.49 9.47 6.88
Year 2 25.18 43.90 32.28 32.42 10.83 15.76 13.27 12.76 6.00 8.34 7.74 6.74
Year 3 1.48 1.00 -0.38 5.61 11.99 15.40 15.16 11.06 15.25 18.97 18.44 10.82
Year 4 10.19 16.76 9.12 8.45 10.06 14.39 14.21 11.17 6.46 10.27 12.59 10.05
Year 5 3.24 7.27 6.09 3.36 9.75 15.22 15.10 12.79 5.52 8.63 7.91 6.90
18
Year 6 2.80 5.66 2.17 1.73 8.16 15.37 11.02 9.76 8.59 16.94 12.02 10.50
Year 7 6.87 15.71 12.55 10.18 8.00 15.87 13.96 11.34 3.65 10.10 8.31 6.40
Year 8 26.45 35.74 30.91 27.68 14.67 15.89 12.28 11.72 13.19 9.96 8.15 7.74
Year 9 -0.51 0.04 2.32 1.08 12.64 14.34 13.48 9.89 11.60 12.19 11.84 9.09
Year 10 12.43 19.71 16.94 15.92 12.73 15.14 14.21 12.53 12.35 12.45 13.26 10.95
Year 1–10 8.84 14.36 10.99 10.65 11.14 15.28 13.54 11.35 15.25 18.97 18.44 10.95
Table 1: Performance data for the EURUSD original strategy and the volatility and EVT based strategies.
5 CONCLUSION
95% VaR (%) 95% CVaR (%) Sharpe Ratio
Strategy Orig. Vol. EVT CDaR Orig. Vol. EVT CDaR Orig. Vol. EVT CDaR
Year 1 1.04 1.50 1.14 0.96 1.47 2.09 1.69 1.30 0.21 0.21 0.05 0.20
Year 2 1.03 1.39 1.28 1.21 1.48 2.10 1.85 1.77 2.33 2.79 2.43 2.54
Year 3 1.23 1.64 1.57 1.13 1.52 1.95 1.97 1.46 0.12 0.07 -0.02 0.51
Year 4 1.01 1.32 1.46 1.17 1.21 1.67 1.82 1.42 1.01 1.17 0.64 0.76
Year 5 1.02 1.54 1.54 1.25 1.39 2.07 2.14 1.86 0.33 0.48 0.40 0.26
19
Year 6 0.69 1.40 0.93 0.81 1.14 2.11 1.57 1.39 0.34 0.37 0.20 0.18
Year 7 0.80 1.62 1.32 1.16 1.29 2.44 2.27 1.84 0.86 0.99 0.90 0.90
Year 8 1.39 1.66 1.16 1.17 2.28 2.17 1.71 1.60 1.80 2.25 2.52 2.36
Year 9 1.32 1.49 1.28 0.96 1.70 1.97 1.89 1.38 -0.04 0.00 0.17 0.11
Year 10 1.09 1.42 1.31 1.20 1.77 2.12 2.01 1.75 0.98 1.30 1.19 1.27
Year 1–10 1.08 1.50 1.33 1.11 1.58 2.10 1.94 1.61 0.79 0.94 0.81 0.94
Table 2: Performance data for the EURUSD original strategy and the volatility and EVT based strategies.
5 CONCLUSION
Return (%) Volatility (%) Max. Drawdown (%)
Strategy Orig. Vol. EVT CDaR Orig. Vol. EVT CDaR Orig. Vol. EVT CDaR
Year 1 -5.40 -3.42 -3.08 -3.13 13.35 14.82 11.89 9.39 14.04 13.95 12.01 9.80
Year 2 43.16 49.21 48.81 32.68 15.23 15.90 16.67 12.82 8.07 8.03 8.60 8.13
Year 3 2.72 -0.68 0.12 -0.41 14.85 15.63 15.24 12.42 10.45 13.17 12.19 9.73
Year 4 -10.06 -9.16 -10.87 -7.59 12.17 14.12 13.19 7.66 14.38 14.74 16.51 10.29
Year 5 -0.93 -1.71 -2.60 -1.40 11.25 15.78 15.37 9.01 13.35 17.17 17.20 10.86
20
Year 6 4.62 6.72 4.84 2.82 12.69 15.22 14.60 9.96 7.21 8.26 7.96 5.55
Year 7 5.57 6.30 7.90 4.83 12.76 15.79 14.54 10.49 13.59 15.52 14.61 10.56
Year 8 11.08 16.52 16.58 9.02 18.11 16.32 17.31 10.24 14.86 8.45 10.31 6.71
Year 9 -3.86 1.54 -5.64 -2.94 17.39 14.62 9.73 5.79 17.40 12.96 10.95 6.83
Year 10 -1.67 -3.04 -3.18 -1.42 10.56 15.47 12.28 7.29 8.82 13.15 11.41 6.67
Year 1–10 3.44 5.04 4.07 2.60 14.03 15.37 14.26 9.73 23.51 24.29 24.79 18.25
Table 3: Performance data for the NZDMXN original strategy and the volatility and EVT based strategies.
5 CONCLUSION
95% VaR (%) 95% CVaR (%) Sharpe Ratio
Strategy Orig. Vol. EVT CDaR Orig. Vol. EVT CDaR Orig. Vol. EVT CDaR
Year 1 1.33 1.38 1.14 0.93 1.77 1.88 1.52 1.23 -0.40 -0.23 -0.26 -0.33
Year 2 1.28 1.32 1.39 1.15 1.79 1.89 1.96 1.55 2.83 3.10 2.93 2.55
Year 3 1.54 1.57 1.55 1.28 1.94 2.22 2.13 1.74 0.18 -0.04 0.01 -0.03
Year 4 1.37 1.48 1.46 0.89 1.65 1.84 1.83 1.05 -0.83 -0.65 -0.82 -0.99
Year 5 1.20 1.63 1.66 0.94 1.51 2.14 2.08 1.21 -0.08 -0.11 -0.17 -0.16
21
Year 6 1.21 1.44 1.35 0.91 1.56 1.88 1.82 1.26 0.36 0.44 0.33 0.28
Year 7 1.36 1.65 1.48 1.05 1.81 2.30 2.06 1.52 0.44 0.40 0.54 0.46
Year 8 1.57 1.48 1.42 0.85 2.58 2.12 2.37 1.41 0.61 1.01 0.96 0.88
Year 9 1.78 1.50 1.08 0.67 2.66 2.09 1.49 0.91 -0.22 0.11 -0.58 -0.51
Year 10 1.13 1.68 1.32 0.80 1.42 2.05 1.72 1.04 -0.16 -0.20 -0.26 -0.19
Year 1–10 1.42 1.52 1.38 0.95 1.93 2.07 1.94 1.35 0.25 0.33 0.29 0.27
Table 4: Performance data for the NZDMXN original strategy and the volatility and EVT based strategies.
5 CONCLUSION
REFERENCES REFERENCES
References
A LGOET, P. H. and C OVER, T. M. (1988). Asymptotic optimality and asymptotic equipartition properties
of log-optimum investment. Annals of Probability, 16(2):876–898.
A NDERSON, J. A. and FAFF, R. W. (2004). Maximising futures returns using fixed fraction asset allocation.
Applied Financial Economics, 14:1067–1073.
A RTZNER, P., D ELBAEN, F., E BER, J. M., and H EATH, D. (1999). Coherent measures of risk. Mathematical
Finance, 9(3):203–228.
BALI, T. G. (2003). An extreme value approach to estimating volatility and value at risk. Journal of
Business, 76(1):83–108.
BALKEMA, A. and DE H AAN, L. (1974). Residual life time at great age. Annals of Probability, 2:792–804.
BARONE -A DESI, G., G IANNOPOULOS, K., and VOSPER, L. (1999). VaR without correlations for portfolios
of derivative securities. Journal of Futures Markets, 19(5):583–602.
B EDER, T. S. (1995). VaR: Seductive but dangerous. Financial Analysts Journal, 51(5):12–24.
B EIRLANT, J., G OEGEBEUR, Y., S EGERS, J., and T EUGELS, J. (2004). Statistics of Extremes: Theory and
Applications. Wiley.
B ERNOULLI, D. (1738). Specimen theoriae novae de mensura sortis. Commentarii Academiae Scientiarum
Imperialis Petropolitannae, Tomus V:175–192.
B ERNOULLI, D. (1954). Exposition of a new theory on the measurement of risk. Econometrica, 22(1):23–
36.
B LACK, F. (1976). Studies in stock price volatility changes. Proceedings of the 1976 American Statistical
Association, Business and Economic Statistics Section., pages 177–181.
B OUDOUKH, J., R ICHARDSON, M., and W HITELAW, R. (1998). The best of both worlds: A hybrid ap-
proach to calculating value at risk. Risk, 11(5):64–67.
B REIMAN, L. (1961). Optimal gambling systems for favorable games. In N EYMAN, J., editor, Proceedings
of the Fourth Berkeley Symposium on Mathematical Statistics and Probability: Contributions to the The-
ory of Statistics, volume 1, pages 65–78. Statistical Laboratory of the University of California, Berkeley,
University of California Press.
B ROWNE, S. and W HITT, W. (1996). Portfolio choice and the Bayesian Kelly criterion. Advances in Applied
Probability, 28(4):1145–1176.
22
REFERENCES REFERENCES
C ASCON, A. and S HADWICK, W. F. (2009). A new approach to tail risk. Journal of Investment Consulting,
10(1):33–48.
C HEKHLOV, A., U RYASEV, S., and Z ABARANKIN, M. (2003). Portfolio Optimization With Drawdown
Constraints, chapter 13, pages 253–268. Asset and Liability Management Tools. Risk Books.
C HEKHLOV, A., U RYASEV, S., and Z ABARANKIN, M. (2005). Drawdown measure in portfolio optimiza-
tion. International Journal of Theoretical and Applied Finance, 8(1):13–58.
C VITANIC, J. and K ARATZAS, I. (1995). On portfolio optimization under ”drawdown” constraints. IMA
Lecture Notes in Mathematics & Applications, 65:77–88.
D UFFIE, D. and PAN, J. (1997). An overview of value at risk. Journal of Derivatives, 4(3):7–49.
E NGLE, R. F. (1982). Autoregressive conditional heteroskedasticity with estimates of the variance of United
Kingdom inflation. Econometrica, 50(4):987–1007.
E NGLE, R. F. (2001). GARCH 101: The use of ARCH/GARCH models in applied econometrics. Journal
of Economic Perspectives, 15(4):157–168.
G HORBEL, A. and T RABELSI, A. (2008). Predictive performance of conditional Extreme Value Theory in
Value-at-Risk estimation. International Journal of Monetary Economics and Finance, 1(2):121–148.
G HORBEL, A. and T RABELSI, A. (2009). Measure of financial risk using conditional extreme value copulas
with EVT margins. Journal of Risk, 11(4):51–85.
G OLDBERG, L. R., H AYES, M. Y., M ENCHERO, J., and M ITRA, I. (2009). Extreme Risk Analysis. MSCI
Working Paper.
G OLDBERG, L. R., M ILLER, G., and W EINSTEIN, J. (2008). Beyond value-at-risk: forecasting portfolio
loss at multiple horizons. Journal of Investment Management, 6(2):73–98.
G ROSSMAN, S. J. and Z HOU, Z. (1993). Optimal investment strategies for controlling drawdowns. Mathe-
matical Finance, 3(3):241–276.
H AKANSSON, N. (1970). Optimal investment and consumption strategies under risk for a class of utility
functions. Econometrica, 38(5):587–607.
23
REFERENCES REFERENCES
H ARDING, D., NAKOU, G., and N EJJAR, A. (2003). The pros and cons of drawdown as a statistical measure
for risk in investments. AIMA Journal, pages 16–17.
H ARMANTZIS, F. C., M IAO, L., and C HIEN, Y. (2006). Empirical study of value-at-risk and expected
shortfall models with heavy tails. Journal of Risk Finance, 7(2):117–135.
H UANG, W., L IU, Q., R HEE, S. G., and W U, F. (2012). Extreme downside risk and expected stock returns.
Journal of Banking & Finance, 36(5):1492–1502.
J OHANSEN, A. and S ORNETTE, D. (2001). Large stock market price drawdowns are outliers. Journal of
Risk, 4(2):69–110.
K ELLY, J. L. (1956). A new interpretation of information rate. Bell System Technical Journal, 35(4):917–
926.
L AJBCYGIER, P. and L IM, E. (2007). How important is money management? Comparing the largest ex-
pected equity drawdown, optimal-f and two naı̈ve money management approaches. Journal of Trading,
2(3):58–75.
L ATAN É, H. A. (1959). Criteria for choice among risky ventures. Journal of Political Economy, 67(2):144–
155.
L OCKE, P. R. and M ANN, S. C. (2003). Prior outcomes and risky choices by professional traders. Working
paper.
L ONGIN, F. M. (2000). From value at risk to stress testing: the extreme value approach. Journal of Banking
& Finance, 24(7):1097–1130.
M AC L EAN, L. C., S ANEGRE, R., Z HAO, Y., and Z IEMBA, W. T. (2004). Capital growth with security.
Journal of Economic Dynamics & Control, 28(4):937–954.
M AC L EAN, L. C., T HORP, E. O., Z HAO, Y., and Z IEMBA, W. T. (2011a). How does the Fortune’s Formula
Kelly capital growth model perform? Journal of Portfolio Management, 37(4):96–111.
M AC L EAN, L. C., T HORP, E. O., and Z IEMBA, W. T. (2010). Good and bad properties of the Kelly criterion.
Working paper.
M AC L EAN, L. C., T HORP, E. O., and Z IEMBA, W. T. (2011b). The Kelly Capital Growth Investment
Criterion: Theory and Practice, volume 3 of Handbook in Financial Economic Series. World Scientific.
M C N EIL, A. J. and F REY, R. (2000). Estimation of tail-related risk measures for heteroscedastic financial
time series: an extreme value approach. Journal of Empirical Finance, 7:271–300.
24
REFERENCES REFERENCES
M C N EIL, A. J., F REY, R., and E MBRECHTS, P. (2005). Quantitative Risk Management : Concepts, Tech-
niques, and Tools. Princeton University Press.
M ENDES, B. V. M. and B RANDI, V. (2004). Modeling drawdowns and drawups in financial markets. Journal
of Risk, 6(3):53–69.
M ENDES, B. V. M. and L EAL, R. P. C. (2005). Maximum drawdown: Models and applications. Journal of
Alternative Investments, 7(4):83–91.
M ETROPOLIS, N. and U LAM, S. (1949). The Monte Carlo method. Journal of the American Statistical
Association, 44(247):335–341.
N ELSON, D. B. and C AO, C. Q. (1992). Inequality constraints in the univariate GARCH model. Journal of
Business and Economic Statistics, 10:229–235.
N YSTR ÖM, K. and S KOGLUND, J. (2005). Efficient filtering of financial time series and Extreme Value
Theory. Journal of Risk, 7(2):63–84.
O SORIO, R. (2009). Prospect-theory approach to the Kelly criterion for fat-tail portfolios: The case of
Student’s t-distribution. Wilmott Journal, 1(2):101–107.
P ICKANDS, J. (1975). Statistical inference using extreme order statistics. Annals of Statistics, 3:119–131.
P RITSKER, M. (2006). The hidden dangers of historical simulation. Journal of Banking and Finance,
30(2):561–582.
ROTANDO, L. M. and T HORP, E. O. (1992). The Kelly criterion and the stock market. American Mathe-
matical Monthly, 99(10):922–931.
ROY, A. D. (1952). Safety first and the holding of assets. Econometrica, 20(3):431–449.
S AMUELSON, P. A. (1971). The “fallacy” of maximizing the geometric mean in long sequences of investing
or gambling. Proceedings of the National Academy of Sciences, 68(10):2493–2496.
S AMUELSON, P. A. (1979). Why we should not make mean log of wealth big though years to act are long.
Journal of Banking & Finance, 3(4):305–307.
T HALER, R. H. and J OHNSON, E. J. (1990). Gambling with the house money and trying to break even: The
effects of prior outcomes on risky choice. Management Science, 36(6):643–660.
T HORP, E. O. (1971). Portfolio choice and the Kelly criterion. Business and Economics Statistics Section,
Proceedings of the American Statistical Association, pages 215–224.
25
REFERENCES REFERENCES
T HORP, E. O. (2006). The Kelly criterion in blackjack, sports betting and the stock market, volume 1 of
Handbook of asset and liability management, chapter 9, pages 385–429. Elsevier.
V INCE, R. (2011). Optimal f and the Kelly criterion. IFTA Journal, 11:21–28.
26
REFERENCES REFERENCES
DR ISSAM STRUB: Dr Strub is a senior member of the Cambridge Strategy research group where he
works on quantitative strategies as well as asset allocation and risk management tools; he has authored a
number of research articles in financial and scientific journals and has been an invited speaker at financial
conferences and roundtables. Prior to joining the Cambridge Strategy, Dr Strub was a graduate student at the
University of California, Berkeley, where he conducted research in an array of fields ranging from Partial
Differential Equations and Fluid Mechanics to Scientific Computing and Optimisation; he obtained a Ph.D.
in Engineering from the University of California in 2009.
Disclaimer: Some services are not available to private inexperienced investors. Services may also not be available to certain investors due to regulatory or other constraints either in the UK or
elsewhere. The information contained herein is not targeted at the residents of any particular country and is not intended for distribution to, or use by, any person in any country or jurisdiction where
such distribution or use would be contrary to local law or regulatory requirements. Information contained herein may be subject to change without notice. Investments in securities for financial
instruments (which include contracts for differences, futures, options spot and forward foreign exchange and off-exchange contracts) can fluctuate in value and you should be aware that you may
not realise the initial amount invested and may incur additional liabilities. As investments in securities or financial instruments may entail above average risks you should carefully consider whether
your financial circumstances permit you to invest and if necessary seek the advice of an independent financial adviser. Foreign exchange denominated securities and financial instruments are subject
to fluctuations in exchange rates that may have a positive or negative effect on the value, price or income derived from the securities or financial instrument concerned. Past performance is not a
reliable indicator of future performance. You are advised that The Cambridge Strategy (Asset Management) Limited is unable to provide advice as to tax consequences of a particular investment
or investment strategy and you are advised to seek professional advice in this respect. This document is issued by the Cambridge Strategy (Asset Management) Limited and is not a solicitation
or instruction to invest. It is provided for information purposes only. The Cambridge Strategy (Asset Management) Limited is authorised and regulated by the Financial Services Authority and
registered with the SEC (US) and the SFA (HK). The Cambridge Strategy (Asset Management) Limited is exempt from the requirement to hold an Australian financial services licence under the
Corporations Act 2001 (Cth) (Class Order 03/1099) in respect of the provision of financial services. It is regulated by the Financial Services Authority (FSA) under UK laws, which differ from
Australian laws. Australian Investors: These materials are provided by a representative of The Cambridge Strategy (Asset Management) Limited (’Cambridge’) and is intended for wholesale
clients as defined in the Corporations Act 2001 (Cth). The information in this presentation is current unless stated otherwise and Cambridge is not under any obligation to update the information
to the extent that it is or becomes out of date or incorrect. It is confidential and has been prepared by Cambridge solely for use in connection with its Programmes. This information must not be
made available, published or distributed to any third party without the prior consent of Cambridge. This information has been prepared without taking into account anyone’s objectives, financial
situation or needs so before acting on it each person should consider its appropriateness to their circumstances before making any investment decision. In particular, a person should consider the
Programme’s investment objectives, risks, fees and other charges. Each person should carefully read and consider any offer documentation before making an investment decision. The information in
this presentation is indicative and may change with market fluctuations. It does not purport to be a comprehensive statement or description of any markets or securities referred to within. Cambridge
assumes no fiduciary responsibility or liability for any consequences financial or otherwise arising from any reliance on this information. Each person should make their own appraisal of the risks
and should consult to the extent necessary, their own legal, financial, tax, accounting and other professional advisors in this respect to any investment in the Programmes. This document is not a
solicitation or instruction to invest. It is provided for information purposes only.
27