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

Re: Portfolio simulation



PureBytes Links

Trading Reference Links

The biggest problem I found in combining systems is that the big winners 
are often taken by different systems and they overlap.  When you combine 
them into one system, you eliminate many "duplicate" trades.


At 08:32 AM 12/14/01 -0500, you wrote:
>While I'm in the process of evaulating portfolio software, I thought I would
>attempt to simulate a portfolio of S&P Systems trading 1 contract by grouping
>all by various system signals into one system.  These systems test
>historically to have around 80 -85% winning percentage.  When I grouped these
>signals together, I've found that the overall winning % drops to around 70%
>and the total profits are much less than if I had viewed each system
>seperately.  The portfolio system is set up to accept multiple entry signals
>and pyramiding, etc. Each entry and exit signal has it's own name to
>distinguish them from one another.  My question is: How is TS looking at
>these signals and why doesn't the portfolio system match up with the
>individual systems results.  These systems trade from both the long and short
>side.  I'm sure there's a glaring flaw in my approach, and I would like
>someone to point it out to me.  thanks......John D....jdev02@xxxxxxx

Bill Brower
Email: 1000mileman@xxxxxxxxxxxxxx
Web Site: insideedge.net