Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : Systems, Strategies and Resources for Trading Futures -- Ignore unavailable to you. Want to Upgrade?

To: Patrick Slevin who wrote (7595)	10/31/1998 9:37:00 PM
From: F Robert Simms	Read Replies (1) \| Respond to of 44573

I will be away from any computer tomorrow. If you have any questions tomorrow, maybe somebody can answer on the Neural Net thread or one of the people who have an email address at the BioComp web site. Best Wishes, Bob

To: Patrick Slevin who wrote (7595)	11/2/1998 8:44:00 AM
From: Tom Trader	Respond to of 44573

>>The short answer is: use as much data as possible to get as many trades as possible to build as much confidence as possible, but don't use so much data that the market behavour is different at the trading end than it was at the training end<< I would agree with the excerpt that you posted I approach the testing of systems by using a 10 year period and then evaluating it based on the middle 4 years -- to define whether the system seems viable, including any optimization of parameters, etc. Thereafter, I apply the model to the first 3 years and the last 3 years of the 10 year period. If the system works well over all of these periods then I view it as worthy of more detailed review. The latter 3 year period must perform at least as well as the former periods and preferably better. But if the latter period does very well but the former period does not--and is marginal or loses money-- the system needs more work or is unstable the way that I look at it. The subsequent review of a potentially viable system looks at the individual trades and evaluates those for things that may skew peformance, draw-downs, etc.