Neural Nets - A tool for the 90's 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 : Neural Nets - A tool for the 90's

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: LastShadow who wrote (618)	8/7/1999 7:55:00 PM
From: Larry Livingston	Read Replies (1) of 871

Interesting article by Steve Ward. I was wondering whether the six to thirty input maximum that Steve mentioned includes Lags. My understanding is that using several lags is useful, but this would mean going way over the maximum. I use a lot of binary inputs to identify specific patterns. 0 or 1 and so already am way above the maximum. When I use lags , I'm already up to a couple of hundred inputs. Can anyone comment who has had any luck using this many inputs. Is it true that 30 inputs including lags is the max? I've always assumed that adding inputs just meant gathering more rows of data to prevent overfitting. I suppose one way around this would be breaking it up into several nets, however how do you know which inputs work best with which others? Could you use a genetic algorithm to sort through them all. What software would be the best. I currently use Neuroshell 2 and Gene Hunter. Unfortunately there is no way to interface the two.

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next