DIVERSEDIVERSE
Illuminating the World of
presented by
HEDGE FUNDSHEDGE FUNDSHEDGE FUNDSAlfred Winslow Jones
The Institutional Investor Journal, August 1968
BRIEF HISTORY of HEDGE FUNDSBRIEF HISTORY of HEDGE FUNDSBRIEF HISTORY of HEDGE FUNDS
The Most Popular
HEDGE FUNDSTRATEGIESHEDGE FUNDSTRATEGIESHEDGE FUNDSTRATEGIES
Where is All the Money inCTAs & MANAGEDFUTURES FUNDSCTAs & MANAGEDFUTURES FUNDSCTAs & MANAGEDFUTURES FUNDS
UNDERSTANDING MOMENTS DISTRIBUTIONUNDERSTANDING MOMENTS DISTRIBUTIONUNDERSTANDING MOMENTS DISTRIBUTION
2.3172.3172.317
WEALTHWEALTHWEALTH
29%15%12%
12%
10%
6%6%
6%4%
Long/Short Equity
CTA/Managed Futures
Emerging Markets
Global Macro
Fixed Income
Event Driven
Relative Value
Multi Strategy
Other
$127
Switzerland
$89
Luxembourg
$34
Sweden
$27
Bermuda
$483
UK
$88
Brazil
NorthAmerica
UniversityEndowments
IndividualInvestors
Fire, Police, Teacher& Other Pensions
Municipalities
Funds ofHedge Funds
6,178Europe
4,092
Africa
113
Australia
165
South America
1,267Asia
637
$49
Netherlands
$34
Germany
$1124
USA
$66
France
Total reported Assets Under Management (AUM) of hedge funds and funds of hedge funds at the end of �rst half 2012. If broken down into individual U.S. dollar bills and placed side by side, they would circle the equator 9,017 times or go to the moon and back 470 times.
Keeping a Portfolio ProtectedIMAGINING THEUNIMAGINABLEIMAGINING THEUNIMAGINABLEIMAGINING THEUNIMAGINABLE
Traditional risk statistics assume that events to the left and right of the mean are equally likely to occur.
Value at Risk (VaR) is the maximum loss a fund can expect within a speci�ed holding period using a speci�ed con�dence level. It is calculated using Traditional Risk Statistics based on a Normal Distribution,
with No Skew, and Mesokurtic Kurtosis.
The traditional VaR model has come under criticism because, in the real world, events can't be easily modeled on a simple, symmetrical curve (normal distribution). Newer, Fat-Tail VaR models better capture the fact that highly improbable and damaging events do occur, and can occur more frequently than traditional risk statistics assume.
Nassim Nicholas Taleb popularized the use of the term “black swan” for these highly improbable events.
size of the assetsunder managementin billions of USD based on management company location
$ trillion
Hedge Fund Investors are
DIVERSE
CONTINENTCONTINENTHedge Funds are on Every
CONTINENT
*Except for Antarctica
“ The logic of the idea
was very clear. It was a
hedge against the vagaries of
the market. You can buy more
good stocks without taking
as much risk as someone
who merely buys. ”
is Relatively Concentrated
IT'S ALL GREEK TO MEIT'S ALL GREEK TO MEIT'S ALL GREEK TO ME Some Common Investment Statistics
3 4 FUNDS3 4 FUNDS3 4 FUNDS
$
are denominated in the US Dollar or Euro
out of
Regulation of U.S. securities industry begins with the Securities Act of 1933. Regulation D exempts companies selling to
“accredited investors” from registering.
1933
Regulation of U.S. funds begins with passage of the Investment
Company Act of 1940.
1940
Fortune magazine publishes article on Alfred
Winslow Jones’ hedge fund - number of hedge funds
grows in the following years.
1966
Alternative Investment Management Association (AIMA) formed to represent all practitioners
in the alternative investment management industry.
1990
SEC adopted rules requiring registered advisors with
$150 million or more in AUM to report comprehensive
fund information.
2011
Securities Exchange Act of 1934 establishes the
Securities and Exchange Commission (SEC).
1934
Managed Funds Association (MFA)
is formed to advocate on behalf of hedge funds and managed
futures �rms.
1991
Alfred Winslow Jones forms �rst hedge fund (coining them “hedged funds”) and eventually
introduces the 20% performance fee.
1949
JOBS Act passed. Allows hedge funds to market themselves to a broader audience.
2012
Rothschild Family introduces the world’s
�rst fund ofhedge funds.
1969 2002
“UCITS III” Directives released in Europe allowing
UCITS to employ alternative strategies.
Financial Services Authority (FSA)
established to regulate the UK �nancial
services industry.
Committee of European Securities Regulators (CESR) established – later
replaced by the European Securities
and Markets Authority (ESMA) in 2011.
2001
56%
USDollar
21%
Euro
7%
Brazilian Real
4%
Chinese Yuan
4%
UKPound
PERCENT OF TOTAL AUM BY FUND SIZE
NUMBER OF HEDGE FUNDS AND FUNDS OF HEDGE FUNDS REPORTING A GIVEN DOMICILE
3%
SwissFranc
5%
Other
SYMBOL DEFINITION WHAT IT ATTEMPTS TO ANSWER
*
< $25M
$25-50M
$51-100M
$101-250M
$251-500M
$501M-1B
>$1B78.1%
1.1%
1.1%
1.8%
4.7%
5.0%
8.2%
Family Of�ces
Foundations & CharitableOrganizations
Measures the fund’s value relative to a benchmark.
Alpha
Beta
Sigma(Standard Deviation)
α
0.135%
β
Ω
σ
Omega
How much extra did you earn from a fund that you wouldn’t have otherwise earned from investing in the broad market?
Measures the fund’s sensitivity to movements of the market as a whole
How likely is your fund to track the benchmark?
Uses actual return distribution and divides expected returns into gains and losses to provide a relative measure of the likelihood of the fund achieving a given return.
How do your fund’s good returns stack up to its bad returns, and how likely is it that your fund will make more than a given percentage?
Measures the degree of variation of the fund’s returns around the fund’s mean (average) return for a speci�ed period.
How much should you expect your fund’s returns to vary from the norm?
THE FORMULA FOR YOUR INNER MATH GEEK
LEPTOKURTIC
For example, traditional risk approaches assume that the
likelihood of an extreme event (negative or positive) is very slim.
These extreme events are modeled by the tails of the
normal distribution.
MESOKURTIC PLATYKURTIC
TRADITIONAL RISK STATISTICS
chance
The overstated probability that an extremely positive event (gain) occurs according to the
traditional risk approach.
Traditional ETL
Fat-Tail ETL
ExtremeNegative Event
(Loss) Occurs Here
An asymmetrical curve, based on real-world returns, can more accurately predict the chance
that an extreme negative event (loss) or an extreme positive event (gain) may occur.
Notice that the tail is “fatter” to the left of 95% VaR. This indicates that the probability of sustaining losses has increased.
chance
ExtremePositive Event
(Gain) Occurs Here
In this example,both the Fat-Tail and
normal distributions havethe same mean and VaR 95%
numbers, but the Fat-Tail distribution is Leptokurtic and
has a left skew.
As a result, the Fat-Tail distribution’s Expected Tail
Loss (ETL), which is the average of returns that exceed the VaR,
more accurately captures a higher downside risk than
traditional ETL.
FAT-TAIL RISK STATISTICS
VARIANCE
MEAN
STANDARD DEVIATIONmeasures the volatility of returns from the mean
SEMI DEVIATIONmeasures the volatility of returns below the mean
LOSS DEVIATIONmeasures the volatility of returns from the mean only during periodsof a loss
DOWNSIDE DEVIATION (5%)measures the volatility of returns below a Minimum Acceptable Return (MAR), offered here at 5%
Where M = The mean return of the benchmarkWhere M = The mean return of the fund
Alpha = M - Beta x M
R
R
RD
RD
2
I-1I 1R I R
N
I-1
N
R
RD
I
IWhere R = The return of the benchmark for period IWhere RD = The return of the fund for period IWhere M = The mean return of the benchmarkWhere M = The mean return of the fundWhere N = Number of periods
Beta = ( ∑ ( R - M )(RD - M ) ) ÷ ( ∑ ( R - M )RD
1/2
1/2
2
I-1I R
N
R
IWhere R = Return for period IWhere M = Mean of return set RWhere N = Number of periods
Standard Deviation = ( ∑ ( R - M ) ÷ (N - 1) )
I=1IR
N
M = ( ∑ R ) ÷ N
Annualized Standard Deviation = Monthly Standard Deviation x ( 12 )
Where r is the threshold return, and F is cumulative density function of returns.
b
r
Ω(r) =
∫F(x)dx
∫(1-F(x))dx
r
a
If You've Made it This Far... It's Time for Some More Advanced Analysis
of athe
Moment
1st
Moment
2nd
Moment
3rd
Moment
4th
Measures the distance of returns from the mean. In other words, variance shows how frequently possible returns occur.
SKEWNESS
NEGATIVE SKEWNO SKEW
Characterizes the degree of asymmetry of a distribution around its mean. In other words, by using skewness investors should be able to better predict whether a return is more likely to occur to the left or to the right of the mean.
KURTOSISCharacterizes the relative peakedness or �atness of a distribution. In other words, kurtosis helps investors better predict the likelihood of a given return (or loss).
The average of returns. In other words, it's the average.
μ
μ = Mean
x
y
x
y
x
y
POSITIVE SKEW
x
y
μ
σ = Standard Deviation
Mode
μMedianMode
(high peak)
(returns more likely to be to the left of the mean)
(returns more likely to be to the right of the mean)
(returns equally likely to be to the left and to the right
of the mean)
(�at topped)(normal distribution)
Median
Mode
Median
μ μ
x
y
x
y
x
y
-1σ-2σ-3σ 1σ 2σ 3σx
y
"One single observation can invalidate a general statement derived from millennia
of confirmatory sightings of millions of white swans. All you need is one single
(and, I am told, quite ugly) black bird.”
Nassim Nicholas TalebThe Black Swan: The Impact of the Highly Improbable
Office locations and telephone numbers:
© 2012 PerTrac
www.pertrac.com
pertrac.com/blog Hong Kong: +1 852.2526.9780
Tokyo: +81.3.5785.2041
London: +44 (0)20. 7651.0800
Memphis: +1 901.888.7300
New York: +1 212.661.6050
Reno: +1 775.851.2282
@PerTrac
facebook.com/pertrac
SovereignWealth Funds
-3σ μ-1σ-2σ 1σ 2σ 3σ
Commodity Futures Trading Commission
(CFTC) established. Commodity Trading Advisors (CTAs), or commodity pools,
are generally required to register with the CFTC.
1974
European Federation of Investment Funds and
Companies (FEFSI) established – later
renamed European Fund and Asset
Management Association (EFAMA).
founded.
1996
Hedge Fund Association (HFA)
founded to advocate on behalf of smaller and
emerging hedge funds.
0.135%
95%
VaR
Images are for illustration purposes and are not to scale
GOOD THINGS COME IN SMALL PACKAGESGOOD THINGS COME IN SMALL PACKAGESGOOD THINGS COME IN SMALL PACKAGESSmaller Funds Tend to Outperform Larger Funds… …But Larger Funds are Generally Less Volatile
CUMULATIVE PERFORMANCE BY SIZE OF FUNDJanuary 1996 to December 2011
ANNUALIZED VOLATILITY BY SIZE OF FUNDJanuary 1996 to December 2011
100%
6.92%
7.39%
5.94%
4.75%4.07%
4.00%
3.63%3.67%
4.12%
5.95%
6.05%5.96%
0% 1% 2% 3% 4% 5% 6% 7% 8%
0%
DEC1996
DEC1998
DEC2000
DEC2002
DEC2004
DEC2006
DEC2008
LARGE> $500M AUM
MID-SIZE$100 - 500M AUM
SMALL< $100M AUM
DEC2010
300%
500%
700%
DOWNSIDEDEVIATION
(5%)
LOSSDEVIATION
SEMIDEVIATION
STANDARDDEVIATION558%
356%
307%
