[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

The Kelly Criterion



PureBytes Links

Trading Reference Links

<x-html><!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>

<META content=text/html;charset=iso-8859-1 http-equiv=Content-Type>
<META content='"MSHTML 4.72.3110.7"' name=GENERATOR>
</HEAD>
<BODY bgColor=#c0c0c0>
<DIV><FONT color=#800000 size=2>Re:&nbsp; <STRONG><U>The Kelly 
Criterion</U></STRONG></FONT></DIV>
<DIV><FONT color=#800000 size=2><STRONG><U></U></STRONG></FONT>&nbsp;</DIV>
<DIV><FONT color=#800000 size=2>Walt,</FONT></DIV>
<DIV><FONT color=#800000 size=2></FONT>&nbsp;</DIV>
<DIV>
<DIV><FONT color=#800000 size=2>In 1730, logarithmic utility was first 
introduced by Daniel Bernoulli.&nbsp; (Todhunter, I., <EM>A History of the 
Mathematical Theory of Probability, </EM>1st edition, Cambridge, 1865, as 
reprinted by Chelsea, New York, 1965).</FONT></DIV>
<DIV><FONT color=#800000 size=2></FONT><FONT color=#800000 
size=2></FONT>&nbsp;</DIV>
<DIV><FONT color=#800000 size=2>In 1956, logarithmic utility was distinguished 
by its properties from other utilities when </FONT><FONT color=#800000 
size=2>John L. Kelly, Jr., an engineer with the Bell Telephone</FONT></DIV>
<DIV><FONT color=#800000 size=2>Company, published his research on efficient 
transmission of electrical signals.&nbsp; (Kelly, J.L., &quot;A New 
Interpretation of Information Rate&quot;, <EM>Bell System Technical 
Journal&nbsp; </EM>(July 1956), 917-926<EM>.</EM>)</FONT></DIV></DIV>
<DIV><FONT color=#800000 size=2></FONT>&nbsp;</DIV>
<DIV><FONT color=#800000 size=2>In 1960, Brieman expanded its application.&nbsp; 
(Brieman, L., &quot;Investment Policies for Expanding Business Optimal in a Long 
Run Sense&quot;, <EM>Naval Research Logistics Quarterly </EM>7 : 4 (1960), 
647-651;&nbsp; Brieman, L., &quot;Optimal Gambling Systems for Favorable 
Games&quot;, <EM>Symposium on Probability and Statistics 4th, Berkley, 1961, 
</EM>1 pp. 65-78).</FONT></DIV>
<DIV><FONT color=#800000 size=2></FONT>&nbsp;</DIV>
<DIV><FONT color=#800000 size=2>In 1971,&nbsp; Edward Thorp used the Kelly 
Criterion to develop investment strategies.&nbsp; (&quot;Portfolio Choice and 
the Kelly Criterion&quot; first published in the 1971 Business and Economics 
Statistics Section Proceedings of the American Statistical Association, and 
later reprinted in the book <EM>Stochastic Optimization Models in Finance,</EM> 
edited by William T. Ziemba and Raymond G. Vickson. New York:&nbsp; Academic 
Press, 1975, pages 599-619.)</FONT></DIV>
<DIV><FONT color=#800000 size=2></FONT>&nbsp;</DIV>
<DIV><FONT color=#800000 size=2>regards,</FONT></DIV>
<DIV><FONT color=#800000 size=2></FONT>&nbsp;</DIV>
<DIV><FONT color=#800000 size=2>Jeff Stewart</FONT></DIV>
<DIV><FONT color=#800000 size=2>Atlanta, Georgia</FONT></DIV>
<DIV><FONT color=#800000 size=2><EM></EM></FONT>&nbsp;</DIV>
<DIV><FONT color=#000000 
size=2>----------------------------------------------------------------------------------------------------------------------------------------</FONT></DIV>
<DIV>The Kelly criterion</DIV>
<DIV><FONT face=Arial size=2>-----Original Message-----<BR>From: Walt Downs 
&lt;knight@xxxxxxxxxxxx&gt;<BR>To: 
RealTraders Discussion Group &lt;<A 
href="mailto:realtraders@xxxxxxxxxxxxxx";>realtraders@xxxxxxxxxxxxxx</A>&gt;<BR>Date: 
Saturday, January 09, 1999 1:17 PM<BR>Subject: Trading GEN: Info on Kelly 
Principle<BR><BR></DIV></FONT>&gt;Hi all,<BR>&gt;<BR>&gt;A few month's back some 
RT's were discussing the basics of the<BR>&gt;&quot;Kelly Principle&quot;, a 
form of statistical adjustment that might<BR>&gt;have value in relation to trade 
size.<BR>&gt;<BR>&gt;I am interested in the original source of this principle. 
I.e.,<BR>&gt;the full name of the author and what book this principal 
is<BR>&gt;discussed in.<BR>&gt;<BR>&gt;Thanks,<BR>&gt;Walt Downs<BR>&gt;CIS 
Trading<BR>&gt;<BR>&gt;</BODY></HTML>
</x-html>From ???@??? Sat Jan 09 13:36:40 1999
Received: from list.listserver.com (198.68.191.15)
	by mail05.rapidsite.net (RS ver 1.0.2) with SMTP id 6769
	for <neal@xxxxxxxxxxxxx>; Sat,  9 Jan 1999 15:50:21 -0500 (EST)
Received: from localhost (localhost [127.0.0.1])
	by accessone.com (8.8.5/8.8.5/PIH) with SMTP id MAA05017;
	Sat, 9 Jan 1999 12:42:59 -0800 (PST)
Received: from laplaza.org (laplaza.org [204.151.72.2])
	by accessone.com (8.8.5/8.8.5/PIH) with ESMTP id MAA04802
	for <realtraders@xxxxxxxxxxxxxx>; Sat, 9 Jan 1999 12:40:26 -0800 (PST)
Received: from laplaza.org (ppp04.questa.laplaza.org [204.151.75.250])
	by laplaza.org (8.9.0/8.9.0) with ESMTP id NAA21010;
	Sat, 9 Jan 1999 13:34:08 -0700 (MST)
Message-Id: <3697BE98.42385F5E@xxxxxxxxxxx>
Date: Sat, 09 Jan 1999 14:39:52 -0600
Reply-To: tlwyss@xxxxxxxxxxx
Sender: owner-realtraders@xxxxxxxxxxxxxx
From: Terry Wyss <tlwyss@xxxxxxxxxxx>
To: RealTraders Discussion Group <realtraders@xxxxxxxxxxxxxx>
Subject: Re: Accutrader -- The Truth vs Bashing
References: <004101be3c05$733f8bc0$8d6840ce@xxxxxxx>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-To: bshumake <bshumake@xxxxxxxxxxxx>, Realtraders <realtraders@xxxxxxxxxxxxxx>
X-Mailer: Mozilla 4.05 [en] (WinNT; I)
X-Listprocessor-Version: 8.1 -- ListProcessor(tm) by CREN
X-Loop-Detect: 1
X-UIDL: 6913a0d3687ba5f6fa2b28178a8b8c9b

Bill,
    Thank you kindly for this well thought out
encouragement to those
of us attempting to learn how to scalp the S&P
consistently. I'm about
a break-even trader right now, and have invested
considerable time, effort
and money to learn how to do so profitably and
consistently. I have no
inclination to quit now and, encouragement and ideas
from those who are
successful is certainly what I need right now, not
misguided though well
meaning pontificating from inexperienced or position
traders, or sour grapes
from failed quitters. I have a sign on my office wall
that has the encouraging
admonishment: "Achieve the Impossible!" and I fully
intend to do just that.
    Wishing you prosperous trading on your own terms,
Terry


bshumake wrote:

>  I have been reading with amusement the various posts
> warning people away from George Heffernan's
> Accutrader.  The amusing parts are the speculation
> that one cannot trade the S&P with 1 point stops and
> strive to make profits in the 1 and 2 point range.
> The part about  not being able to have 80%+
> returnswas also pretty good.  While I,  like the
> people criticizing Accutrader, have not traded the
> system, I do know that the claims are feasible.  I
> day trade the S&P using 2 minute charts ( accutrader
> uses 3 or 5 minute charts ) so it is definitely
> possible to trade that short of a time frame.  My
> accuracy while not 85%, is currently 79.6% which is
> not far removed,  I am sure there are some cynics
> that would tell people 79% is not possible.  Lastly
> I never risk more that 1 1/2 points on a trade, and
> more often than not, risk only 1  point per trade. I
> have been consistently profitable since I began
> trading this method, which is a mechanical, momentum
> based system that uses simple off-the-shelf
> indicators and one price pattern.  Also if you ask me
> what my slippage is I will tell you it is zero...the
> reason is,  I am scalping to pick up profits between
> 1 and 2 points.  My target which uses a limit order (
> no slippage with those ) is always between 1 and 2
> points from where I enter the market ( not from where
> my entry stop is ),  My commissions are $20.oo per
> round turn.  Please don't buy into the notion the
> trades must be docked $200 or even $100 when looking
> at a track record, that can be drastically affected
> by the type of order used as well as where the order
> is placed.  I suspect the idea of $100-$200 slippage
> originated from some classic trading techniques that
> would have you place an order in the same area where
> hundreds of other orders are placed. Also please
> don't begrudge someone for selling their system as
> there are some good reasons for doing so.  It could
> be the trader has developed a decent system, but has
> no capital or is under capitalized ( I did call
> George Heffernan about his system and this is the
> reason he gave for selling it. )  Another good reason
> for selling a system is that it is not compatible
> with your personality.  I can attest that day trading
> is much more stressful than position trading ( for me
> anyway ).  So if a position trader develops a good
> day trading system, trys it for a while but finds it
> too stressful, he may decide to sell it and return to
> position trading.  This doesn't mean that the system
> was not good or did not make money, it simply means
> that it was not a good fit for that particular
> trader...it could be a marriage made in heaven for
> someone else.   I am a good example of this, as I
> have been position trading for many years, but have
> been interested in diversifying over multiple time
> frames so developed the daytrading methodology I am
> currently using.  While it is proving to be very
> profitable it is also proving to be a very intense
> form of trading and more stressful than I
> anticipated...in other words, while I continue to
> trade it, I am not yet convinced that it is a good
> match for my personality ( I am currently hoping my
> personality will adjust somewhat, given time ).
> Trading, aside from being profitable, must also be
> done in a way that is enjoyable.  Life is too short
> to have it any other way. It is relatively easy to
> develop a profitable system.  It is much harder to
> implement that system in real trading due to our
> human emotions of fear and greed ( But where would
> the markets be without fear and greed?).There are
> some very good system developers in the world such as
> Tom DeMark, who do not trade because they do not have
> the disposition for it, but instead opt to sell their
> services to others.  DeMark for example, was on Tudor
> Jones' payroll for several years.  Here is an example
> of a great trader, Jones, buying systems from a
> developer, DeMark. Lastly, do not assume that because
> someone is profitable in a daytrading or scalping
> environment that they would automatically be
> profitable in a money management or hedge fund
> environment.  I know that my own system, while very
> profitable trading small lots ( 5 contracts is
> probably the upper limit ), is simply not a feasible
> approach when trading larger lots.  The result would
> simply be partially filled orders, which would
> quickly erode profits down to zero.  To trade
> hundreds of millions or often even millions of
> dollars takes a much longer term approach than Mr.
> heffernan's system provides.  So even if it is as
> profitable as he claims  ( and it could be ), there
> is a good reason he is not snatched up by a hedge
> fund and made a super star trader...his system simply
> is inappropriate for that scale of trading and would
> not work.  This is analogous to saying that a
> profitable floor trader could be placed in an
> off-floor environment  and still be profitable using
> his old floor trading methods.  He wouldn't be, he
> would have to adopt new methods or go broke. In
> closing, let me urge you not to be cynical about good
> claims.  Certainly there are charlatans and cheats
> around, just as in any business, you often have to
> sift through the crap to find the nuggets.  In
> accutraders case, a written money back guarantee is
> offered.  Also when ordering trading books,systems,
> etc, just use a credit card and you can always have
> the charge revoked ( I have done this on one occasion
> ).Also with regards to cynicism, remember there are
> plenty of "respectable" financial advisor types that
> when asked about someone making money in futures (
> say Richard Dennis, Tudor Jones, or Linda Raschke )
> will tell the questioner ( with a bit of a haughty
> laugh ) that it is simply not possible, that " if it
> sounds to good to be true it probably is."   As a
> result, their client walks away from what might have
> been a very rewarding and profitable pursuit.. All
> the Best !Bill Shumake