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: Optim who wrote (670)	3/28/2000 12:12:00 AM
From: xu, b.	Read Replies (1) of 871

>FWIW, I find it much easier to model on a weekly basis. >Some of the best nets I have ever developed for the S&P >are weekly traders. >Optim I would like to thank Lastshadow for bringing me here. It's an informative forum and has been a little bit dead recently. Just want to offer my 2cents here: By using weekly data, you already use a filter to filter out the daily fluctuation. You will have a cleaner data set for your NN to understand. NN is nothing but a brain function modeling tool and it's a very primitive one so it needs some helps. I am in a buisness to make structural simulation of human brain which is much more complex. Having said that, I do believe ANN is a very powerful tool and it's so powerful that one can easily fall into curve fitting if one is not careful. I have little knowledge in ANN development, however, my data analysis and trading system develpment experiences seem to suggest that if an NN is only good for one stock or one future contract, it's probably a dangerous beast. Robustness test is one the keys for a good trading system. Bai

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next