This action might not be possible to undo. Are you sure you want to continue?

Column 24

Understanding The Kelly Criterion – Part II

by Edward O. Thorp

Copyright 2008

During a recent interview in the Wall Street Journal (March 22-23,

2008) Bill Gross and I discussed turbulence in the markets, hedge funds and

risk management. Bill considered the question of risk management after he

read Beat the Dealer in 1966. That summer he was off to Las Vegas to beat

blackjack. Just as I did some years earlier, he sized his bets in proportion to

his advantage, following the Kelly Criterion as described in Beat the Dealer,

and ran his $200 bankroll up to $10,000 over the summer. Bill has gone from

managing risk for his tiny bankroll to managing risk for Pacific Investment

Management Company’s (PIMCO) investment pool of almost $1 trillion. He

still applies lessons he learned from the Kelly Criterion. As Bill said, “Here at

PIMCO it doesn’t matter how much you have, whether it’s $200 or $1 trillion.

… Professional blackjack is being played in this trading room from the

standpoint of risk management and that’s a big part of our success.”

The Kelly Criterion applies to multiperiod investing and we can get

some insights by comparing it with Markowitz’s standard portfolio theory for

single period investing.

Compound Growth and Mean–Variance Optimality

Nobel Prize winner Harry Markowitz introduced the idea of mean-

variance optimal portfolios. This class is defined by the property that, among

the set of admissible portfolios, no other portfolio has both higher mean

return and lower variance. The set of such portfolios is known as the efficient

frontier. The concept is a cornerstone of modern portfolio theory, and the

mean and variance refer to one period arithmetic returns. In contrast the

Kelly Criterion is used to maximize the long term compound rate of growth, a

multiperiod problem. It seems natural, then to ask the question: is there an

analog to the Markowitz efficient frontier for multiperiod growth rates, i.e. are

there portfolios such that no other portfolio has both a higher expected

growth rate and a lower variance in the growth rate? We’ll call the set of

such portfolios the compound growth mean-variance efficient frontier.

Let’s explore this in the simple setting of repeated independent

identically distributed return per unit invested, the random variables

{ ¦ : 1, ,

i

X i n · K with ( ) 0

i

E X > so the “game” is favorable, and where the non-

negative fractions bet at each trial are specified in advance { ¦ : 1, , .

i

f i n · K To

keep the math simpler, we also assume that the

i

X are not constant and

have a finite number of distinct values. After

n

trials the compound growth

6/25/2010 1

rate per period is { ¦ ( ) ( )

1

1

log 1

n

i i i

i

G f f X

n

·

· +

∑

and the expected growth rate

{ ¦ ( ) { ¦ ( ) ( ) ( ) ( )

1 1

1 1

log 1 log 1 log 1 .

n n

i i i i i

i i

g f E G f E f X E f X E fX

n n

· ·

]

· · + · + ≤ +

]

∑ ∑

The

last step follows from the (strict) concavity of the log function, X has the

common distribution of the ,

i

X we define

1

1

n

i

i

f f

n

·

·

∑

and we have equality if

and only if

i

f f · for all . i Therefore if some

i

f differ from f we have

{ ¦ ( ) { ¦ ( )

.

i

g f g f < This tells us that betting the same fixed fraction always

produces a higher expected growth rate than betting a varying fraction with

the same average value. Note that whatever f turns out to be, it can always

be written as

*

, f cf · a fraction

c

of the Kelly fraction.

Now consider the variance of { ¦ ( )

.

i

G f If X is a random variable with

( ) ( ) , 1 , P X a p P X q · · · − · and 0 a > then

( )

2

1

ln 1 ln .

1

af

Var fX pq

f

] | ` +

] + ·

]

]

−

. , ]

(Compare Thorp, 2007, section 3.1). Note: the change of variable

, 0, f bh b · > shows the results apply to any two valued random variable. We

chose 1 b · for convenience.) A calculation shows that the second derivative

with respect to f is strictly positive for 0 1 f < < so ( ) ln 1 Var fX ] +

]

is strictly

convex in . f It follows that

{ ¦ ( ) ( ) ( ) ( )

1 1

1 1

log 1 log 1 log 1

n n

i i i i

i i

Var G f Var f X Var f X Var fX

n n

· ·

] ] ] ] · + · + ≥ +

] ] ] ]

∑ ∑

with equality if and only if

i

f f · for all . i Since every admissible strategy is

therefore “dominated” by a fractional Kelly strategy it follows that the mean-

variance efficient frontier for compound growth is a subset of the fractional

Kelly strategies. If we now examine the set of fractional Kelly strategies

{ ¦ { ¦

*

f cf · we see that for 0 1, c ≤ ≤ both the mean and the variance increase

as

c

increases but for 1, c ≥ the mean decreases and the variance increases

as

c

increases. Consequently { ¦

*

f dominates the strategies for which 1 c >

and they are not part of the efficient frontier. No fractional Kelly strategy is

dominated for 0 1. c ≤ ≤ We have established in this limited setting:

Theorem. For repeated independent trials of a two valued random

variable, the mean-variance efficient frontier for compound growth over a

finite number of trials consists precisely of the fractional Kelly strategies

{ ¦

*

: 0 1 . cf c ≤ ≤

So, given any admissible strategy, there is a fractional Kelly strategy

with 0 1 c ≤ ≤ which has a growth rate that is no lower and a variance of the

growth rate that is no higher. The fractional Kelly strategies in this instance

are preferable in this sense to all the other admissible strategies, regardless

6/25/2010 2

of any utility function upon which they may be based. This deals with yet

another objection to the fractional Kelly strategies, namely that there is a

very wide spread in the distribution of wealth levels as the number of periods

increases. In fact this eventually enormous dispersion is simply the

magnifying effect of compound growth on small differences in growth rate

and we have shown in the theorem that in the two outcome setting this

dispersion is minimized by the fractional Kelly strategies. Note that in this

simple setting, a utility function will choose a constant

i

f cf · which will

either be a fractional Kelly with 1 c ≤ in the efficient frontier or will be too

risky, with 1, c > and not be in the efficient frontier.

As a second example suppose we have a lognormal diffusion process

with instantaneous drift rate

m

and variance

2

s

where as before the

admissible strategies are to specify a set of fixed fractions { ¦

i

f for each of

n

unit time periods, 1, , . i n · K Then for a given f and unit time period

( )

2 2

VarG f s f · as noted in (Thorp, 2007, eqn. (7.3)). Over

n

periods

{ ¦ ( )

2 2 2 2

1 1

n n

i i

i i

VarG f s f s f

· ·

· ≥

∑ ∑

with equality if and only if

i

f f · for all . i This

follows from the strict convexity of the function ( )

2

. h x x · So the theorem

also is true in this setting. I don’t currently know how generally the convexity

of ( ) ln 1 Var fX + is true (I suspect rather broadly) but whenever it is, and we

also have ( ) ln 1 Var fX + increasing in , f then the compound growth mean

variance efficient frontier is once again the set of fractional Kelly strategies

with 0 1. c ≤ ≤

Samuelson’s Criticisms

The best known “opponent” of the Kelly Criterion is Nobel Prize winning

economist Paul Samuelson, who has written numerous polemics, both

published and private, over the last 40 years. William Poundstone’s book

Fortune’s Formula gives an extensive account with references. The gist of it

seems to be:

(1) Some authors once made the error of claiming, or seeming to

claim, that acting to maximize the expected growth rate (i.e. logarithmic

utility) would approximately maximize the expected value of any other

continuous concave utility (the “false corollary”).

Response: Samuelson’s point was correct but, to me and others,

obvious the first time I saw the false claim. However, the fact that some

writers made mistakes has no bearing on an objective evaluation of the

merits of the criterion. So this is of no further relevance.

(2) In private correspondence Samuelson has offered examples and

calculations in which he demonstrates, with a two valued X (“stock”) and

three utilities, ( ) 1/ , H W W · − ( ) log , K W W · and ( )

1 2

, T W W · that if any one

who values his wealth with one of these utilities uses one of the other utilities

to choose how much to invest then he will suffer a loss as measured with his

6/25/2010 3

own utility in each period and the sum of these losses will tend to infinity as

the number of periods increases.

Response: Samuelson’s computations are simply instances of the

following general fact proven 30 years earlier by E. Thorp and R. Whitley, in

“Concave Utilities are Distinguished by Their Optimal Strategies,” Colloquia

Mathematica Societatis Janos Bolyai 9. “Progress in Statistics,” Proceedings

of the European Meeting of Statisticians, Budapest, 1972. Edited by J. Grani,

S. Sarkadi, and I. Vincze. North Holland, 1974, pp. 813-830.

Theorem 1. Let U and V be utilities defined and differentiable on

( ) 0, ∞ with ( ) U x ′

and ( ) V x ′

positive and strictly decreasing as x increases.

Then if U and V are inequivalent, there is a one period investment setting

such that U and V have distinct sets of optimal strategies. Furthermore, the

investment setting may be chosen to consist only of cash and a two-valued

random investment, in which case the optimal strategies are unique.

Corollary 2. If the utilities U and V have the same (sets of) optimal

strategies for each finite sequence of investment settings, then U and V are

equivalent.

Two utilities

1

U and

2

U are equivalent if and only if there are

constants a and b such that ( ) ( ) ( )

2 1

0 , U x aU x b a · + > otherwise

1

U and

2

U are inequivalent.

Thus no utility in the class described in the theorem either dominates

or is dominated by any other member of the class.

Samuelson offers us utilities without any indication as to how we ought

to choose among them, except perhaps for this hint. He says that he and an

apparent majority of the investment community believe that maximizing

( ) 1/ U x x · − explains the data better than maximizing ( ) log . U x x · How is it

related to fractional Kelly? Does this matter? Here are two examples which

show that this utility can choose

*

cf for any 0 1, c < <

1

2

, c ≠ depending on the

setting.

For a favorable coin toss and ( ) 1/ U x x · − we have

( )

2

*

/ f p q µ · +

which increases from / 2 µ or half Kelly to

µ

or full Kelly as

p

increases from

½ to 1, giving us the set

1

2

1. c < < On the other hand if

( ) ( )

1

2

1 P X A P X · · · − · describe the returns and 1 A > so 0, µ > ( ) 1 / 2 A µ · −

and the Kelly

*

/ . f A µ · For ( ) 1/ U x x · − we find U maximized for

( )

( )

{ ¦

( )

1/ 2

2

2

2 4 1 / 1 f A A A A A A · − t + − − which is asymptotic to

1/ 2

A

−

as A

increases, compared to the Kelly

*

f which is asymptotic to 1/ 2 as A

increases, giving us the set

1

2

0 . c < <

In the continuous case the relation between ( ) , c g f and ( ) ( )

G f σ is

simple and the tradeoff between growth and spread in growth rate as we

adjust

c

between 0 and 1 is easy to compute and it’s easy to visualize the

correspondence between fractional Kelly and the compound growth mean-

6/25/2010 4

variance efficient frontier. This is not the case for these two examples so the

fact that ( ) 1/ U x x · − can choose any

1

2

, 0 1, c c c < < ≠ doesn’t necessarily

make it undesirable. I suggest that a useful way to look at the problem for

any specific example involving

n

period compound growth is to map the

admissible portfolios into the { ¦ ( ) { ¦ ( ) ( )

,

i i

G f g f σ plane, analogous to the

Markowitz one period mapping into the (standard deviation, return) plane.

Then examine the efficient frontier and decide what tradeoff of growth

versus variability of growth you like. Professor Tom Cover points out that

there is no need to invoke utilities. Adopting this point of view, we’re simply

interested in portfolios on the compound growth efficient frontier whether or

not any of them happen to be generated by utilities. The Samuelson

preoccupation with utilities becomes irrelevant. The Kelly or maximum

growth portfolio, which as it happens can be computed using the utility

( ) log , U x x · has the distinction of being at the extreme high end of the

efficient frontier.

For another perspective on Samuelson’s objections, consider the three

concepts normative, descriptive and prescriptive. A normative utility or other

recipe tells us what portfolio we “ought” to choose, such as “bet according to

log utility to maximize your own good.” Samuelson has indicated that he

wants to stop people from being deceived by such a pitch. I completely

agree with him on this point. My view is instead prescriptive: how to achieve

an objective. If you know future payoff for certain and want to maximize your

long term growth rate then Kelly does it. If, as is usually the case, you only

have estimates of future payoffs and want to come close to maximizing your

long term growth rate, then to avoid damage from inadvertently betting more

than Kelly you need to back off from your estimate of full Kelly and consider a

fractional Kelly strategy. In any case, you may not like the large drawdowns

that occur with Kelly fractions over ½ and may be well advised to choose

lower values. The long term growth investor can construct the compound

growth efficient frontier and choose his most desirable geometric growth

Markowitz type combinations.

Samuelson also says that ( ) 1/ U x x · − seems roughly consistent with

the data. That is descriptive, i.e. an assertion about what people actually do.

We don’t argue with that claim – it’s something to be determined by

experimental economists and its correctness or lack thereof has no bearing

on the prescriptive recipe for growth maximizing.

I met with economist Oscar Morgenstern (1902-1977), coauthor with

John von Neumann of the great book, The Theory of Games and Economic

Behavior, at his company “Mathematica” in Princeton, New Jersey, in

November of 1967 and, when I outlined these views on normative,

prescriptive and descriptive he liked them so much he asked if he could

incorporate them into an article he was writing at the time. He also gave me

an autographed copy of his book, On the Accuracy of Economic Observations,

which has an honored place in my library today and which remains timely.

(For instance, think about how the government has made successive

revisions in the method of calculating inflation so as to produce lower

numbers, thereby gaining political and budgetary benefits.)

6/25/2010 5

Proebsting’s Paradox

Next, we look at a curious paradox. Recall that one property of the

Kelly Criterion is that if capital is infinitely divisible, arbitrarily small bets are

allowed, and the bettor can choose to bet only on favorable situations, then

the Kelly bettor can never be ruined absolutely (capital equals zero) or

asymptotically (capital tends to zero with positive probability). Here’s an

example that seems to flatly contradict this property. The Kelly bettor can

make a series of favorable bets yet be (asymptotically) ruined! Here’s the

email discussion through which I learned of this.

From: Todd Proebsting

Subject: FW: incremental Kelly Criterion

Dear Dr. Thorp,

I have tried to digest much of your writings on applying the Kelly Criterion to

gambling but I have found a simple question that is unaddressed. I hope you

find it interesting:

Suppose that you believe an event will occur with 50% probability and

somebody offers you 2:1 odds. Kelly would tell you to bet 25% of your capital.

Similarly, if you were offered 5:1 odds, Kelly would tell you to bet 40%.

Now, suppose that these events occur in sequence. You are offered 2:1 odds,

and you place a 25% bet. Then another party offers you 5:1 odds. I assume

you should place an additional bet, but for what amount?

If you have any guidance or references on this question, I would appreciate it.

Thank you.

From: Ed Thorp

To: Todd Proebsting

Subject: Fw: incremental Kelly Criterion

Interesting.

After the first bet the situation is:

A win gives a wealth relative of 1 + 0.25*2

A loss gives a wealth relative of 1 - 0.25

Now bet an additional fraction f at 5:1 odds and we have:

A win gives a wealth relative of 1 + 0.25*2 + 5f

A loss gives a wealth relative of 1 - 0.25 - f

The exponential rate of growth g(f) = 0.5*ln(1.5+5f) + 0.5*ln(0.75-f)

Solving g'(f) = 0 yields f = 0.225 which was a bit of a surprise until I thought

about it for a while and looked at other related situations.

From: Todd Proebsting

To: Ed Thorp

Subject: RE: incremental Kelly Criterion

Thank you very much for the reply.

I, too, came to this result, but I thought it must be wrong since this tells me to

bet a total of 0.475 (0.25+0.225) at odds that are on average worse than 5:1,

and yet at 5:1, Kelly would say to bet only 0.400.

Do you have an intuitive explanation for this paradox?

From: Ed Thorp

To: Todd Proebsting

Subject: Re: incremental Kelly Criterion

I don't know if this helps, but consider the example:

6/25/2010 6

A fair coin will be tossed (Pr Heads = Pr Tails = 0.5). You place a bet which

gives a wealth relative of 1+u if you win and 1-d if you lose (u and d are both

nonnegative). (No assumption about whether you should have made the

bet.) Then you are offered odds of 5:1 on any additional bet you care to

make. Now the wealth relatives are, each with Pr 0.5, 1+u+5f and 1-d-f. The

Kelly fraction is f = (4-u-5d)/10. It seems strange that increasing either u or d

reduces f. To see why it happens, look at the ln(1+x) function. This odd

behavior follows from its concave shape.

From: Todd Proebsting

To: Ed Thorp

Subject: RE: incremental Kelly Criterion

Yes, this helps. Thank you.

It is interesting to note that Kelly is often thought to avoid ruin. For instance,

no matter how high the offered odds, Kelly would never have you bet more

than 0.5 of bankroll on a fair coin with one single bet. Things change,

however, when given these string bets. If I keep offering you better and

better odds and you keep applying Kelly, then I can get you to bet an amount

arbitrarily close to your bankroll.

Thus, string bets can seduce people to risking ruin using Kelly. (Granted at

the risk of potentially giant losses by the seductress.)

From: Ed Thorp

To: Todd Proebsting

Subject: Re: incremental Kelly Criterion

Thanks. I hadn't noticed this feature of Kelly (not having looked at string

bets).

To check your point with an example I chose consecutive odds to one of A_n:1

where A_n = 2^n, n = 1,2,... and showed by induction that the amount bet at

each n was f_n = 3^(n-1)/4^n (where ^ is exponentiation and is done before

division or multiplication) and that sum{f_n: n=1,2,...} = 1.

A feature (virtue?) of fractional Kelly strategies, with the multiplier less than 1,

e.g. f = c*f(kelly), 0<c<1, is that it (presumably) avoids this.

In contrast to Proebsting’s example, the property that betting Kelly or

any fixed fraction thereof less than one leads to exponential growth is

typically derived by assuming a series of independent bets or, more

generally, with limitations on the degree of dependence between successive

bets. For example, in blackjack there is weak dependence between the

outcomes of successive deals from the same unreshuffled pack of cards but

zero dependence between different packs of cards, or equivalently between

different shufflings of the same pack. Thus the paradox is a surprise but

doesn’t contradict the Kelly optimal growth property.

My previous column on the Kelly Criterion, in the May 2008 issue, had

an unfortunate number of typographical errors and omitted Table 1. I have

posted a corrected version on my website, www.EdwardOThorp.com.

6/25/2010 7

- Stochastic Calculus - Book
- St. Petersburg Portfolio Games
- Principal Component and Constantly Re-balanced Portfolio - Slides
- On the History of the Growth Optimal Portfolio
- Non Parametric Prediction
- Nonparametric Estimation for Financial Investment Under Log-Utility
- Growth Optimal Portfolio Selection Strategies With Transaction Costs
- Kernel-Based Semi-log-optimal Empirical Portfolio Selection Strategies
- Empirical Log-optimal Portfolio Selections - A Survey
- An Asymptotic Analysis of the Mean-Variance Portfolio Selection
- FIX-5.0_SP1_VOL-7
- FIX-5.0_SP1_VOL-6
- FIX-5.0_SP1_VOL-5
- FIX-5.0_SP1_VOL-4
- FIX-5.0_SP1_VOL-3
- FIX-5.0_SP1_VOL-2
- FIX-5.0_SP1_VOL-1
- Time Series Documentation - Mathematica
- Time Series Analysis Book
- Quantitative Methods in Derivatives Pricing
- Portfolio Formation Can Affect Asset Pricing Tests
- Modelling Financial Time Series With S-Plus - Book
- Fourier Series
- Course Description Financial Price Analysis - Spring 2010
- Analysis of High-Frequency Financial Data With S-Plus

Sign up to vote on this title

UsefulNot useful- Advanced Financial Models
- Column Understanding the Kelly Criterion
- Putting Volatility to Work
- Pairs Trading--Quantitative Methods and Analysis (Wiley Books)
- Understanding Fortunes Formula
- stochint
- blatt11
- Modeling Derivatives in C Tqw Darksiderg
- TilsonBehavioralFinance
- ekonomikos knyga
- Quantitative Methods in Mgmt Ref1
- Number Play - Gyles Brandreth
- Gatarek D. Bachert P. Maksymiuk R. the LIBOR Market Model in Practice
- The Mathematics of Gambling Edward O Thorpe
- The Quants by Scott Patterson - Excerpt
- Rational Choice
- Risk Neutral+Valuation Pricing+and+Hedging+of+Financial+Derivatives
- Quantitative Methods in Derivatives Pricing
- Lecture Notes
- A Map and Simple Heuristic to Detect Fragility,Antifragility, and Model Error NNT
- Complete Newbie s Guide to Online Forex Trading
- Financial Mathematics (Lectures)
- OR
- An introduction to statistical inference and its applications
- 3642125972
- LP_driver
- Rachev S.T. Financial Eco No Metrics (LN, Karlsruhe 2006)(146s)_FL
- Logic gates
- The Future Has Thicker Tails Than The Past
- Curtis Arnold - The PPS Trading System.pdf
- Column 24 Understanding the Kelly Criterion 2