|
PureBytes Links
Trading Reference Links
|
Thank you all who responded, I appreciate your input.
JTraDer
-----Original Message-----
From: VK [mailto:volker@xxxxxxxxxxxxxx]
Sent: Tuesday, January 27, 2004 10:11 AM
To: JTraDer421@xxxxxxxxxxx
Cc: catapult@xxxxxxxxxxxxxxxxxx; camacazi@xxxxxxxxxxx; omega-list@xxxxxxxxxx
Subject: AW: Coding Question on Stridsman
Hi,
I have just finished an interview with Thomas, who happens to be a very
knowledgable and nice guy. You will be able to see his interview on our web
site today or tomorrow. BTW, you can take a look at some version of the
system here:
http://www.wealth-lab.com/cgi-bin/WealthLab.DLL/editsystem?id=13291. The WS
code is there and you can run it across various individual stocks or test it
on a portfolio of symbols for free. The system has been coded in November
2002 for WLD2.1! WLD is doing most of the calculation in percent or at some
point gives the option to look at points or dollar value.
May be it helps and gives you some ideas.
Volker Knapp
(Wealth-Lab Inc.)
||-----Ursprüngliche Nachricht-----
||Von: DH [mailto:catapult@xxxxxxxxxxxxxxxxxx]
||Gesendet: Tuesday, January 27, 2004 2:21 AM
||An: Omega List
||Betreff: Re: Coding Question on Stridsman
||
||
||> If you agree with Stridsman's idea that percentage based performance
||> measures are the way to go, one way to get similar results from a
||> standard TS system test is to use constant dollar (rather than
||> constant
||> contract) position sizing.
||
||Yep, Stridesman has created a complicated solution for a trivial
||problem. Of course, if the answer is 2 lines of TS code, it's hard to
||write and sell a book about that. :-)
||
||--
|| Dennis
||
||
|