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: Optim who wrote (38)	1/7/1998 6:55:00 PM
From: current trend	Read Replies (2) \| Respond to of 871

By running more "neurons" are you increasing computing power -- speed of output, or would your end results be different with variable numbers of neurons? What limits the number of neurons besides software design? CT

To: Optim who wrote (38)	1/7/1998 7:10:00 PM
From: alan von weiler	Read Replies (2) \| Respond to of 871

Hi Optim; Yes I am a Canuck. I have all my data starting from 08/95. I thought that if I upgraded to a GA program with 100 neurons, I would obtain better results, but if you have a confidence level in the 60's, well, I don't know, maybe I should be happy with what I have. My target stock, ICA in Vancouver, is at 35 cents today. My net says that if I traded with NS since 08/95, the start of my historical data, I would have 6 - 7 million dollars now. My problem with stocks that trade so low in price and volume, to sell all ICA, I would have to sell a lot of stock(ie:$7,000,000 divided by 35 cents)and the market could not absorbe the high volume. ICA's volume today was 1000 shares although typically it is in the 25000 plus range. I think NS is better suited for higher volume stocks. That is my guess. But if I can get NS to predict the market price up to 10 days in advance, the sky's the limit. The thing is that NS must prove itself. So time will tell. I thought that if I run 5 different nets with the same target stock but with different comparable stock or index "baskets", that would be a good test. I have analysed results so far and find that nets with lower run times all say to sell. Regards, Alan