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 : Neural Nets - A tool for the 90's -- Ignore unavailable to you. Want to Upgrade?

To: LastShadow who wrote (655)	9/2/1999 4:58:00 PM
From: Optim	Read Replies (2) \| Respond to of 871

Gothca! ;) It seems we miss each other on points at times because we have a slightly different focus. Most of my successful models thus far have been based on intermarket-type inputs (related issues, interest rates, etc.). From what I gather, you are more of an avid 'OHLC contains all' or 'fundamentals are reflected in the price' believer. I have to admit the OHLC approach (if I can call it that!) does have it's advantages as relationships that are found tend to remain constant longer. For example the Open/Low relationship is usually a good input for many securities and it tends to remain so. My problem with the OHLC approach is that I haven't yet made the distinction on how to extract or highlight these influences for the benefit of the nets. I have used a GA to find relationships that do well, but the net tends to shrug them off. Yet if I test these as a simple mechanical system they remain constant and profitable. My focus lately has been on cycles in the OHLC and how to explain the concept of time to my nets. I'm hoping to use something similar to Fibonacci time projections to help the net identify likely turning points. Optim