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: Patrick Slevin who wrote (323)	11/28/1998 12:27:00 PM
From: CatLady	Read Replies (2) \| Respond to of 871

Re: Neural Net Books Old post from Optim to me on NN books - Message 3151898 And the neural network FAQ - ftp://ftp.sas.com/pub/neural/FAQ.html You'll have to cut 'n' paste the FTP link into your browser's location box, it didn't work as http.

To: Patrick Slevin who wrote (323)	11/28/1998 8:19:00 PM
From: Carl R.	Read Replies (1) \| Respond to of 871

I decided to run an experiment. I took a known good model that has performed well out of sample for me and pared down the inputs to a minimum, about 6-8. I know that using more, like 16, causes the CATNN and TDNN models to not generalize well, so I used the smaller model, and since I already knew which inputs were the most important, I eliminated the least significant ones. I selected "make smaller models" and also "use fewer inputs". On all of the above models most used 4 inputs and a few used up to 6. Then I ran NGO selecting only 1 type of model, and ran about 150 models, then looked at the 10 best to see how they performed. I did not make any special selection of models, but rather used exactly what NGO made. I had to leave to buy a christmas tree so the CATNN models ran 500 models, and thus CATNN performed a bit better than the others. Here are my results by model type: $/model $/model models/min In Sample Out of Sample CATNN $80,002 $15,742 1.8 TDNN 85,127 12,933 2.5 BP 70,439 6,688 8 SOM 73,026 7,997 15 GRNN 189,964 (1,463) 17 The in-sample period was from 7/10/95 to 6/30/98, about 3 years. The out of sample period was 7/1/98 to current, or 5 months. Since the out of sample period is only 1/7 as long as the in-sample period many of the models performed just as well out of sample as in sample. I would conclude that GRNN does a great job of curve fitting, but has no predictive value. If you want to run fast, use BP or SOM, but do not use GRNN. Other than the GRNN models, I don't think that there is a significant difference between the models, so long as there is enough data. I ran PNN also (14/minute) but got only out-of-market models. I started TSOM but it ran extremely slowly and the initial models were all zero. This was a small model, and thus ran quicker by a wide margin than some of the 32 input models I have tried previously. Still a CATNN only run of 10,000 models would take 91 hours on my Cyrix. Carl

To: Patrick Slevin who wrote (323)	11/28/1998 8:28:00 PM
From: Carl R.	Read Replies (1) \| Respond to of 871

Here is a suggestion to help you get better models. Pick something that you normally trade by some other method. Pick a few things that you normally use that you think help you decide when to buy and sell such as put-call ratio, bollinger bands, or whatever. Limit yourself to no more than 6-8 inputs. Choose a variable to predict, optimize your 6 inputs, and let 'er rip. Save some data for out-of-sample testing. If you have the CPU power, go for walk-forwards testing of at least 2-3 periods, but keep a sizable training data set. After 50 iterations, download the models and check them on the data you held back. Unless at least 9 of the top 10 are profitable out-of-sample, start over, otherwise, let it run to completion. Leave NGO on random refill for the first 50 iterations to make sure you have nice diverse models. CATNN includes TDNN with floating time delays, and is thus TDNN is a subset of TDNN. BP includes no time delays and is thus a subset of TDNN (and CATNN). Since they are simpler there are less degrees of freedom and they require less data (or can handle more variables). If these models are looking at the same things you usually do, they should be able to interpret them as well as you can, or better. Hope this helps, Carl