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 : Technical Analysis - Beginners -- Ignore unavailable to you. Want to Upgrade?

To: David R. Evans who wrote (7635)	1/12/1998 4:02:00 AM
From: shasta23	Read Replies (2) \| Respond to of 12039

HI DAVE! Have started looking into system testing. Only have MSWIN though therefore my capabilities are limited what i heard. I neglected system testing when i got MSWIN and concentraded on Explorations and indicators because the results in testing were very inconsistent and didn't seem to make sense at all. Q+ is my provider and i only use the last two years of data. I can almost hear the scream of Richard or other experienced TAer that this is nonsense but i still think i can get first experiences with this. Pulling of data from the MSWIN data CD is timeconsuming if i want to look at multiple chart. And i hope that Q+ soon will bring out the 32-bit version for all of us (and not only the betatester) which then will give us i think 8 years of data. But anyway...you wrote that in the end of the stock selection process you run a systemtest to see which of the selected stock will test best against the selection criterium e.g.:cross(stochrsi(14),30) for buy and cross under 70 for sell. I get often results i'm not satisfied with after running this test. The reasons seem to be: 1)After entering the stock the stochrsi peaks without reaching a value >70 and then begins a downtrend. We will be long the whole time altough we might loose 50% of our capital. The stochrsi(14) might drop under 30 and then up again and we will not get a new buy signal since we're still considered long and can't buy at the new bottom. ONly after we reached finally a value above 70 and then drop below it a real sell signal is given. This is all correct according to the test syntax but will give us a nice loss. In reality we would've left the trade long time ago when we reached more than 10% loss or the 5EMA crossed under 40 EMA etc. So how can you decide which of these stocks test better under these kind of testing conditions that grossly alter the real trading situation? 2)When i looked through my databse based on IBD with 70/70 EPS/RS then i find a lot of stocks that rallied in 1997 out of a long base 1996.For these kind of charts DAHL and DNS seem to work much better. Systems based on Oversold/oberbought like CCI or Stochrsi seemed to trade these stocks but miss a lot of points.The seemed to work better on stocks that go up and down a lot.Do you agree? So right now i'm not very happy with these results because i don't know how to integrate system testing. Yes, i heard the message "test, test, test!" and will do that but if you find time to write something about this part of the stock selection process i would be very grateful. I think you wrote great posts about selection of the right stocks in the past and lately the one about exiting. I also read a nice post about FIB retracement early in this thread. But i haven't seen you talk about system testing...maybe you could/should/would... Thanks for all your help! Stefan