|
PureBytes Links
Trading Reference Links
|
<x-html><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=text/html;charset=iso-8859-1 http-equiv=Content-Type><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"><!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<META content="MSHTML 5.00.2614.3500" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2>M. G.,</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Here is my suggestion. Find or develop a
trading system that you can verify, through backtesting, that has a positive
expectancy of profit. I recommend EOD trading rather than
daytrading. Remember that 100 test cases proving you have a good system
has an uncertainty of 10%. Once you are satisfied you have the system to
trade, develop a trading plan that outlines exactly what you are going to do to
trade this system. There are many books out there that talk to developing
a plan. I recommend Van Tharp's book on trading. </FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>With $10,000 you need to remember that you cannot
risk very much with each trade, so stick with markets that are
appropriate. You are starting out at a big disadvantage with such a small
starting equity. You need to do a lot of analysis to prove that your
system won't make you go bust. I've done it with a small account, but my
system doesn't take big risks.</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>The statistics for futures daytraders is pretty
sobering. A lot wash out. Therefore my recommendation to start out
with End of Day trading. If you can't make a go of EOD, don't move to
daytrading thinking that will do the trick. It will only get
worse!</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>I hope this helps. Just one man's opinion
.. Marlowe</FONT></DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV></BODY></HTML>
</x-html>From ???@??? Thu Jan 20 13:25:54 2000
Return-Path: <listmanager@xxxxxxxxxxxxxxx>
Received: from mail.thetrellis.net ([208.179.56.11])
by purebytes.com (8.8.7/8.8.7) with SMTP id OAA19411
for <neal@xxxxxxxxxxxxx>; Thu, 20 Jan 2000 14:31:14 -0800
Received: from REALTRADERS.COM
([208.179.56.198])
by mail.thetrellis.net; Thu, 20 Jan 2000 13:23:30 -0800
Received: from mailhost2.dircon.co.uk by realtraders.com
with SMTP (MDaemon.v2.8.5.0.R)
for <realtraders@xxxxxxxxxxxxxxx>; Thu, 20 Jan 2000 13:17:52 +0000
Received: from dellpc2 (th-en136-135.pool.dircon.co.uk [194.112.54.135])
by mailhost2.dircon.co.uk (8.9.3/8.9.3) with SMTP id VAA61967;
Thu, 20 Jan 2000 21:20:07 GMT
Reply-To: <r-hodge@xxxxxxxxxxxxxxx>
From: "Robert Hodge" <r-hodge@xxxxxxxxxxxxxxx>
To: <realtraders@xxxxxxxxxxxxxxx>
Subject: [RT] RE: RE: Pre-processing for Neural Nets
Date: Thu, 20 Jan 2000 21:21:41 -0000
Message-ID: <NDBBIAJCHEAJLMPDFAKOAELLCDAA.r-hodge@xxxxxxxxxxxxxxx>
MIME-Version: 1.0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0)
Importance: Normal
In-Reply-To: <01BF62E8.91418C30@xxxxxxxxxxxxxxxxxxx>
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
X-MDaemon-Deliver-To: realtraders@xxxxxxxxxxxxxxx
X-Return-Path: r-hodge@xxxxxxxxxxxxxxx
Sender: listmanager@xxxxxxxxxxxxxxx
X-MDMailing-List: realtraders@xxxxxxxxxxxxxxx
X-MDSend-Notifications-To: listmanager@xxxxxxxxxxxxxxx
Status:
Thanks for the reply Mark.
You wrote:
"You can use wavelet's, but that is more than mere temporal compression, as
wavelet's also perform partial frequency analysis. Do you want to dissect a
signal into various time-localized frequency components? If not, use
vanilla temporal compression."
For me and any others who might be having difficulty with this first
sentence or so of your reply let me say some of it back. It may or may not
make sense after being filtered by my brain... :-)
I can use a wavelet transform to encode say X daily closes in Y wavelet
coefficients each representing the relative strength of a particular wavelet
"size" (ie length in time) at the point where the input ends...Each
coefficient in turn only represents fluctuation in the input at a higher
frequency than the last - ie the same frequency information is not presented
as output by more than one of the coefficients (??)
Whereas, if I used Y streams of the same daily closes sampled at Y different
rates (to notionally achieve the same effect) I would in practice be
presenting similar frequency information but getting some overlap in the
process...In other words its not "wrong" it just presents redundant
frequency information when compression and independence of each input is at
a premium (ie input to a neural net).
Close...or incoherent rambling?
Regards,
Robert
P.S. For general group interest I gleaned from an old email from Bob Brickey
on the Tradelab list (and elsewhere) that a useful book on this stuff is
Timothy Masters' "Neural,Novel and Hybrid Algorithms for Time Series
Prediction". Someone also recommended Murray Ruggeiro's "Cybernetic
Trading...". No doubt Mark Jurik's other book "Neural Networks and
Financial Forecasting" would also be interesting.
|