PureBytes Links
Trading Reference Links
|
I may have introduced, or perhaps identified, a common logical leap into oblivion which is distracting when discussing the random nature of the marketor stocks. If a time series has characteristics of a random distribution,i.e. if its distribution is "close enough" to another distribution, I suggest that may not be of the utmost importance to the trader. The technical analysts is implicitly hoping for is some significance to the time behaviorof the time series, the history in this context. If the next trade is purely random and a Markov process, independent of the previous information, why not simply place your bets? Contrary to popular intuition of the general population, hopefully not of this group, if there have been a dozen consecutive heads from a coin toss - the likelihood of the next toss being headsis --- 50%.
You may find message http://groups.yahoo.com/group/amibroker/message/19978 to be relevant. It displays two curves with identical random distributionshowever one has an obvious history while the other does not. I have reposted the gif to this note.
I suggest we are more interested in the history than the distribution, in most cases.
Cordially,
Richard
Message -----
From: E Winters
To: amibroker@xxxxxxxxxxxxxxx
Sent: Saturday, June 29, 2002 1:19 PM
Subject: Re: [amibroker] Trading system development and simulations
Richard,
You have stated the null hypothesis - that trades occur in a random sequence. To test the alternative hypothesis that their was a pattern to the trades you would calculate the z statistic for the observed sequence of trades and compare the calculated z with the tabulated z for a predetermined level of significance. 1 to 5 percent is good. If the calculated z statisticis outside the expected range then you would have reason to suspect that the sequence of trades was non-random. This of course assumes that the trades follow a binomial distribution curve, which does fit market behavior well, but it does give an idea of how random the trades are.
Regards,
Ed
----- Original Message -----
From: Richard Alford
To: amibroker@xxxxxxxxxxxxxxx
Sent: Saturday, June 29, 2002 10:06 AM
Subject: Re: [amibroker] Trading system development and simulations
If the process is random (as the game likely is) there is no such thingas a "streak", specifically it is a Markov process with no history feedback. It would be interesting to understand just what Tharp's game teaches.
Some think/hope/assume/pray/etc, btw, that there is information in the time history of an security. A "system" that exploits this history and getto conditions which are favorable may indeed have a "streak".
www.bjmath.com has some interesting articles. The BJ is BlackJack, which has well know properties that can tilt the odds towards a player at times.
Cheers
Richard
----- Original Message -----
From: E Winters
To: amibroker@xxxxxxxxxxxxxxx
Sent: Friday, June 28, 2002 6:02 PM
Subject: Re: [amibroker] Trading system development and simulations
David,
Van Tharp has a demo of his trading program on his web site. If you agree to get a free copy of his newsletter and register (he does send announcements about his courses to registered users) you can download it and tryit out. Look for the free trading game box on the main page at www.iitm.com
One theory is based on the theory of runs. When you are on a winningstreak you should risk more per trade until you have a loser. Try adding to winning streaks and see how it improves trading.
Regards,
Ed
----- Original Message -----
From: David Holzgrefe
To: amibroker@xxxxxxxxxxxxxxx
Sent: Friday, June 28, 2002 5:41 PM
Subject: Re: [amibroker] Trading system development and simulations
AL an interesting article , Has anyone tried his trading game ?? what did they learn from it
David
----- Original Message -----
From: Avcinci
To: amibroker@xxxxxxxxxxxxxxx
Sent: Saturday, June 29, 2002 6:32 AM
Subject: [amibroker] Trading system development and simulations
Here is a very interesting article I got from Tharp's forum by Frank Gallucci. It is about trading system development and simulating. It's agood read: http://www.iitm.com/Weekly_update/simulations1.htm
Al Venosa
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.372 / Virus Database: 207 - Release Date: 6/20/2002
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
Yahoo! Groups Sponsor
ADVERTISEMENT
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
------=_NextPart_001_00CB_01C220D7.5BE1B010
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2716.2200" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2>Actually, my comments most directly addressed the
concept of "streaks", mentioned in the referenced post about Tharp's money
game:<EM> If the process is random (as the game likely is) there is no such
thing as a "streak", specifically it is a Markov process with no history
feedback. It would be interesting to understand just what Tharp's game
teaches. </EM></FONT></DIV>
<DIV><EM><FONT face=Arial size=2></FONT></EM> </DIV>
<DIV><FONT face=Arial size=2>I may have introduced, or perhaps identified, a
common logical leap into oblivion which is distracting when discussingthe
random nature of the market or stocks. If a time series has
characteristics of a random distribution, i.e. if its distribution is "close
enough" to another distribution, I suggest that may not be of the utmost
importance to the trader. The technical analysts is implicitly hopingfor
is some significance to the time behavior of the time series, the history in
this context. If the next trade is purely random and a Markov process,
independent of the previous information, why not simply place your bets?
Contrary to popular intuition of the general population, hopefully not of this
group, if there have been a dozen consecutive heads from a coin toss - the
likelihood of the next toss being heads is --- 50%.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>You may find message <A
href="http://groups.yahoo.com/group/amibroker/message/19978">http://groups.yahoo.com/group/amibroker/message/19978</A> to
be relevant. It displays two curves with identical random distributions
however one has an obvious history while the other does not. I have
reposted the gif to this note.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>I suggest we are more interested in the history
than the distribution, in most cases.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Cordially,</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Richard</FONT></DIV>
<DIV> </DIV>
<DIV> Message ----- </DIV>
<BLOCKQUOTE
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV
style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B>
<A title=e.winters@xxxx href="mailto:e.winters@xxxxxxxxxx">E
Winters</A> </DIV>
<DIV style="FONT: 10pt arial"><B>To:</B> <A title=amibroker@xxxxxxxxxx
href="mailto:amibroker@xxxxxxxxxxxxxxx">amibroker@xxxxxxxxxxxxxxx</A> </DIV>
<DIV style="FONT: 10pt arial"><B>Sent:</B> Saturday, June 29, 2002 1:19
PM</DIV>
<DIV style="FONT: 10pt arial"><B>Subject:</B> Re: [amibroker] Trading system
development and simulations</DIV>
<DIV><BR></DIV>
<DIV><FONT face=Arial size=2>Richard,</FONT></DIV>
<DIV><FONT face=Arial size=2>You have stated the null hypothesis - that trades
occur in a random sequence. To test the alternative hypothesis that
their was a pattern to the trades you would calculate the z statistic forthe
observed sequence of trades and compare the calculated z with the tabulated z
for a predetermined level of significance. 1 to 5 percent is good.
If the calculated z statistic is outside the expected range then you would
have reason to suspect that the sequence of trades was non-random. This
of course assumes that the trades follow a binomial distribution curve, which
does fit market behavior well, but it does give an idea of how random the
trades are.</FONT></DIV>
<DIV><FONT face=Arial size=2>Regards,</FONT></DIV>
<DIV><FONT face=Arial size=2>Ed</FONT></DIV>
<BLOCKQUOTE
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV style="FONT: 10pt arial">----- Original Message ----- </DIV>
<DIV
style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B>
<A title=richard.alford@xxxx href="mailto:richard.alford@xxxx">Richard
Alford</A> </DIV>
<DIV style="FONT: 10pt arial"><B>To:</B> <A title=amibroker@xxxxxxxxxxxx
href="mailto:amibroker@xxxxxxxxxxxxxxx">amibroker@xxxxxxxxxxxxxxx</A></DIV>
<DIV style="FONT: 10pt arial"><B>Sent:</B> Saturday, June 29, 2002 10:06
AM</DIV>
<DIV style="FONT: 10pt arial"><B>Subject:</B> Re: [amibroker] Tradingsystem
development and simulations</DIV>
<DIV><BR></DIV>
<DIV><FONT face=Arial size=2>If the process is random (as the game likely
is) there is no such thing as a "streak", specifically it is a Markov
process with no history feedback. It would be interesting to
understand just what Tharp's game teaches.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Some think/hope/assume/pray/etc, btw, that
there is information in the time history of an security. A "system"
that exploits this history and get to conditions which are favorable may
indeed have a "streak". </FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2><A
href="http://www.bjmath.com">www.bjmath.com</A> has some interesting
articles. The BJ is BlackJack, which has well know properties that can
tilt the odds towards a player at times.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Cheers</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Richard</FONT></DIV>
<DIV> </DIV>
<DIV>----- Original Message ----- </DIV>
<BLOCKQUOTE
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV
style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B>
<A title=e.winters@xxxx
href="mailto:e.winters@xxxx">E Winters</A> </DIV>
<DIV style="FONT: 10pt arial"><B>To:</B> <A
title=amibroker@xxxxxxxxxxxxxxx
href="mailto:amibroker@xxxxxxxxxxxxxxx">amibroker@xxxxxxxxxxxxxxx</A>
</DIV>
<DIV style="FONT: 10pt arial"><B>Sent:</B> Friday, June 28, 2002 6:02
PM</DIV>
<DIV style="FONT: 10pt arial"><B>Subject:</B> Re: [amibroker] Trading
system development and simulations</DIV>
<DIV><BR></DIV>
<DIV><FONT face=Arial size=2>David,</FONT></DIV>
<DIV><FONT face=Arial size=2>Van Tharp has a demo of his trading program
on his web site. If you agree to get a free copy of his newsletter
and register (he does send announcements about his courses to registered
users) you can download it and try it out. Look for the free trading
game box on the main page at </FONT><FONT face=Arial size=2><A
href="http://www.iitm.com">www.iitm.com</A></FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>One theory is based on the theory of
runs. When you are on a winning streak you should risk more per
trade until you have a loser. Try adding to winning streaks andsee
how it improves trading.</FONT></DIV>
<DIV><FONT face=Arial size=2>Regards,</FONT></DIV>
<DIV><FONT face=Arial size=2>Ed</FONT></DIV>
<BLOCKQUOTE
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV style="FONT: 10pt arial">----- Original Message ----- </DIV>
<DIV
style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B>
<A title=dtholz@xxxx
href="mailto:dtholz@xxxx">David Holzgrefe</A> </DIV>
<DIV style="FONT: 10pt arial"><B>To:</B> <A
title=amibroker@xxxxxxxxxxxxxxx
href="mailto:amibroker@xxxxxxxxxxxxxxx">amibroker@xxxxxxxxxxxxxxx</A>
</DIV>
<DIV style="FONT: 10pt arial"><B>Sent:</B> Friday, June 28, 2002 5:41
PM</DIV>
<DIV style="FONT: 10pt arial"><B>Subject:</B> Re: [amibroker] Trading
system development and simulations</DIV>
<DIV><BR></DIV>
<DIV><FONT face=Arial size=2>AL an interesting article , Has anyone
tried his trading game ?? what did they learn from it </FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>David</FONT></DIV>
<BLOCKQUOTE
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV style="FONT: 10pt arial">----- Original Message ----- </DIV>
<DIV
style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B>
<A title=avcinci@xxxx
href="mailto:avcinci@xxxx">Avcinci</A> </DIV>
<DIV style="FONT: 10pt arial"><B>To:</B> <A
title=amibroker@xxxxxxxxxxxxxxx
href="mailto:amibroker@xxxxxxxxxxxxxxx">amibroker@xxxxxx</A>
</DIV>
<DIV style="FONT: 10pt arial"><B>Sent:</B> Saturday, June 29, 2002
6:32 AM</DIV>
<DIV style="FONT: 10pt arial"><B>Subject:</B> [amibroker] Trading
system development and simulations</DIV>
<DIV><BR></DIV>
<DIV>Here is a very interesting article I got from Tharp's forum by
Frank Gallucci. It is about trading system development and
simulating. It's a good read: <A
href="http://www.iitm.com/Weekly_update/simulations1.htm">http://www.iitm.com/Weekly_update/simulations1.htm</A></DIV>
<DIV> </DIV>
<DIV><BR>Al Venosa</DIV>
<DIV><BR><TT>Your use of Yahoo! Groups is subject to the <A
href="http://docs.yahoo.com/info/terms/">Yahoo! Terms of
Service</A>.</TT> <BR></DIV><FONT
face="Courier New"></FONT></BLOCKQUOTE>
<BLOCKQUOTE
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px"><FONT
face="Courier New"><BR>---<BR>Outgoing mail is certified Virus
Free.<BR>Checked by AVG anti-virus system (<A
href="http://www.grisoft.com">http://www.grisoft.com</A>).<BR>Version:
6.0.372 / Virus Database: 207 - Release Date:
6/20/2002</FONT></BLOCKQUOTE><BR><TT>Your use of Yahoo! Groups is
subject to the <A href="http://docs.yahoo.com/info/terms/">Yahoo!Terms
of Service</A>.</TT> <BR></BLOCKQUOTE><BR><BR><TT>Your use of Yahoo!
Groups is subject to the <A
href="http://docs.yahoo.com/info/terms/">Yahoo! Terms of Service</A>.</TT>
<BR></BLOCKQUOTE><BR><BR><TT>Your use of Yahoo! Groups is subject to the <A
href="http://docs.yahoo.com/info/terms/">Yahoo! Terms of Service</A>.</TT>
<BR></BLOCKQUOTE><BR><BR><TT>Your
use of Yahoo! Groups is subject to the <A
href="http://docs.yahoo.com/info/terms/">Yahoo! Terms of Service</A>.</TT>
<BR></BLOCKQUOTE></BODY></HTML>
------=_NextPart_001_00CB_01C220D7.5BE1B010--
Attachment:
gif00379.gif
Attachment:
Description: "Description: GIF image"
|