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: Carl R. who wrote (161)	7/6/1998 9:06:00 PM
From: Optim	Read Replies (1) \| Respond to of 871

Actually, I still believe in letting NGO pick its inputs. You can force the NGO to use less inputs and train smaller networks through the genetic control panel. This works well for me, and I use it mainly to find which inputs contain the most information for my models. Then you apply DDR to refine and decorrelate these inputs. I have tried the AMA, CFB, VEL, WAV, and DDR. If you buy them all at once you get a great discount, so I went for the whole lot. The AMA in particular is a great addition to your preprocessing tools. WAV would be the next up in my opinion. One word of caution though. You must be comfortable with Excel and at least have a working knowledge of VBA to get the most from these tools. Once you get you basic sheets set up though it is pretty painless. I use a number of tools, which I have acquired over the course of about 1 1/2 years. My main tools are Metastock (great for OLE and DDE), Excel 97, Neuroshell Trader, BioComp NGO/Profit, and GeneHunter (GA package for Excel). I also do some TA with SuperCharts and I use OmniTrader to screen for candidate stocks. I find that no one tool does everything, and I achieve better results when using combinations. Sorry for the late reply... Optim