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
Neural Nets - A tool for the 90's
An SI Board Since November 1997
Posts SubjectMarks Bans
871 32 0
Emcee:  Optim Type:  Unmoderated
Introduction Compact
Previous 25 | Next 25 | View Recent | Post Message
Go to reply# or date (mm/dd/yy):
ReplyMessage PreviewFromRecsPosted
321 I guess it's probably also wise to include sector activity and broader marPatrick Slevin-11/28/1998
320 <GRNN is quick, but maybe falls apart quickly too.> Why would you thinkPatrick Slevin-11/28/1998
319 Both large and small stocks. I have found that some small caps are predictableCarl R.-11/27/1998
318 I have been using TDNN and GRNN lately. I will have to turn BP back on. Talk abF Robert Simms-11/27/1998
317 What types of stocks do you run this on Carl, large cap Blues and heavy NasdaqPatrick Slevin-11/27/1998
316 My memory of FPU speed was a bit rusty. The PII is only about 15% faster than Carl R.-11/27/1998
315 Because I crash so often, I download top models frequently and save them in proCarl R.-11/26/1998
314 Yes, I came out in '76 as well but I'm older than that because I had tPatrick Slevin-11/26/1998
313 Interesting. I am BSIE 1976, so it sounds as if you are about the same. Of coCarl R.-11/26/1998
312 I printed your post out so I won't forget to go thru it later. At the momPatrick Slevin-11/26/1998
311 Hm, you've just given me an idea. My Bachelor's degree is in IE. Of cPatrick Slevin-11/26/1998
310 BP is back propagation, the model type that runs fairly quickly. The TDNN modelCarl R.-11/26/1998
309 Considering GZ's success, I need to pay more attention to forks. Since you F Robert Simms-11/26/1998
308 Pinnacle sent me a program that eliminates the need for GETDATA. I least I thiPatrick Slevin-11/26/1998
307 I'm not as technically oriented as you are, I'm not quite certain whatPatrick Slevin-11/26/1998
306 The fact that Mike can approach or surpass GZ in the last month is not bad. EspF Robert Simms-11/26/1998
305 <i>I imagine it's quite easy to transpose a number and completely fouCarl R.-11/26/1998
304 <a leak from the future> Yes, I was testing for that by writing down thPatrick Slevin-11/26/1998
303 I do not read the discussion thread at all, hardly. I'm more dismayed thPatrick Slevin-11/26/1998
302 <quick good results.> No, I was not that overwhelmed. I had some good rPatrick Slevin-11/26/1998
301 You wrote: <i>So, in verification and test I might have a short signal goCarl R.-11/25/1998
300 I have data problems with the models that use futures data too. My models that F Robert Simms-11/25/1998
299 Unlike Cook I use real commissions, 35 bucks total. I never count slippage andPatrick Slevin-11/25/1998
298 That is probably a good idea to wait until February before you trust the nets. F Robert Simms-11/25/1998
297 Since the zero cross is somewhat arbitrary at choosing entries and exits the OpF Robert Simms-11/25/1998
Previous 25 | Next 25 | View Recent | Post Message
Go to reply# or date (mm/dd/yy):
HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2026 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News