NeuroStock Message Board

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 : NeuroStock

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: CVDave who wrote (50)	9/17/1998 10:47:00 AM
From: Jay Hartzok	Read Replies (1) of 805

Dave, I numbered my responses according to your numbered paragraphs to keep things coherent. #1 Before I can comment on this, I need to know your definition of a "good model". That SPX net you mentioned sounds to me like it is terribly inaccurate and I'm wondering why you are still training it. #2 Not quite sure of exactly what you are doing on this subject. Please elaborate. I would like more details. #3 Through quite a lot of testing I found that using indices for relateds seems to produce more accurate nets. Most indices are used just once as long term relateds, however if a particular stock's trading patterns closely follow a particular index, I will also use it as a short term related. #4 Anytime that I ever mixed different short or long term influence periods, the net seemed to get confused, so I now maintain consistency. For all inputs I use 20 for short term relateds and normally 3 for long term relateds. As far as price and volume is concerned, if the data is in the file, I usually check the box and let Neurostock sort out the useful information. #5 I use 3 years whenever possible. Jay

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next