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 : Systems, Strategies and Resources for Trading Futures -- Ignore unavailable to you. Want to Upgrade?

To: Tom Trader who wrote (15029)	2/8/1999 6:37:00 PM
From: Brian Hornby	Read Replies (1) \| Respond to of 44573

My first tests used only 3 years of data for "training" and out of sample data were from 8/1/98 to present. I likely will expand both of those. I recall your comments on how bonds may be more predictable, this seems to bear that out. I believe other people using Profit have found the same thing. Certainly the SP systems can go very wrong and out of 10 systems built I typically get some that are losing money. Though note I have just started with this. (edit - which could mean that I get more losing systems as time passes ! <g>) Can you tell me what factors you use in your Bond system? For example. do you look at short and long term interest rates and other variables such as foreign currencies and the price of gold? Also what sort of technical transformations of the data do you feel are useful? Basically for the NN system I do all the transforms, feed in any other data such as foreign currency rates, and then the system finds any correlations and builds a neural net model based on the results. I want to be a little intelligent in what I throw into the soup, as I want to understand how each variable affects the result. I plan to make one system just using TA indicators for the US itself, and another system using everything that could possibly influence bonds, and of course experiments in between. Brian