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 : The Final Frontier - Online Remote Trading -- Ignore unavailable to you. Want to Upgrade?

To: Morpher who wrote (3187)	2/25/1998 7:02:00 PM
From: Eric P	Read Replies (1) \| Respond to of 12617

Morpher: <<Where did you get the 3 weeks of L2 data? Did you record it yourself? What quote system did you use?>> I use a satellite PC Quote datafeed. It is very reliable and very fast. It meets my needs because it sends out the L2 data on all stocks simultaneously, not just the stocks that you have a window open in. In other words, you get ALL of the L2 data and all trades for every stock. Very powerful ==> If you can harness a computer to analyze and make some sense out of the reams of data. Since this data is so crucial for backtesting, I have begun to save each days data. It takes up significant space... I used to only save data on the top ~18 stocks (30 MB/day). Now I am saving all of the data for every stock over $10/share, which consumes ~200 MB of hard disk space per day. My typical routine is to run the 3+ weeks of data overnight against 100 simulated accounts, each consisting of a slightly different permutation of the same general trading system. It takes about 18 hours to go through all of the data on my computer (200 MMX, 128 MB RAM). Then I analyze and record the results and decide what testing I want to run overnight next... One of the things that I analyze for is the time of day that the trade was executed. Last night I did a test run exclusively for short positions for the first time. I would expect the results to be almost entirely independent than the results for the long trades. Interestingly, the best hours for trades were between 1-3 pm EST. Below is the summary, by time, for the 100 accounts: Time # trades Avg $/trade 10-11 1045 +$46 11-12 314 +$212 12-1 424 ($489) 1-2 240 +$337 2-3 333 +$517 3-4 115 ($477) Note: Based on $50/trade commission and SOES executions with 10 second delay at the bid or ask (i.e. losing the spread on every trade). I agree that more data and testing is needed to develop greater confidence in these results. But, it looks interesting for now... I hope this has helped to stimulate some creative thoughts. :-) Happy Trading, -Eric