SI
SI
discoversearch

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 : NeuroStock
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: CatLady who wrote (509)12/27/1998 9:37:00 PM
From: Bob Anderton  Read Replies (2) of 805
 
Feline Frauline, (and Bill)

Thanks very much for all the info. I think I've decided to go ahead a spring for the NS40 commercial based on the input I've received and the impression that seeing the last ten days of trend and volatility bars is necessary to trade with these models. I've only tried two stocks so far, AMAT and ASND, but both are blowing away the best results I've achieved with traditional TA methods for the same periods. I've been training with 1 1/2 years of data and using 7 months of verification data after that to evaluate how the system did later. I'm only using the TXX technology index as the only related (as per Jay and Len's suggestion). I'm still on the free unregistered 6 neuron version so maybe I've gotten lucky in my choice of stocks, or maybe I have yet to appreciate what a truly accurate model can do. It's going to be hard to continue to trade with the tried and true methods that I have experience with while I gain some experience with these NN's. It does seem like the predicted vs actual plots look more like a snowstorm than a diagonal line, but the models are in and out at the right times. The error distribution is a nice tight peak at zero however. What does this mean?

I just exported the AMAT trade report into Excel and for the 7 months of verification data I was getting 1294% annualized. This is for the data that the model was not trained with! Cool.

I'd like to thank all the participants of this thread for sharing their experiences and information. It's made me much more comfortable and apt to push over the learning hump to a whole new set of tools. I hope I can return the favor in the months ahead.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  

HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2025 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News