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


To: Jan Robert Wolansky who wrote (7463)1/2/1998 11:39:00 PM
From: Richard Estes  Respond to of 12039
 
"Shouldn't a system bear the burden of having to prove itself with a wide universe of stocks?"

Dave will bring his answer, but mine is here too. Yes it should face a broad look at all types of stocks. A ranging stock won't return good results using a trending system, and vice versa. But you do want to see a postive results on 80% of stocks tested. You are looking for a system to trust so if PPD turns into a GM, your system won't hurt you.

I recommend a minimum of 5 years data, this length of time usually lets you see how system worked over the different life cycles of a stock. You can find many stocks that were >80 when they were $5 and still there when they are at $50. If someone wants to trade old stogies, they built systems for them. If you want the movers (CANSLIM), you build it for them.

BTW: someone said that they got a 350% return on a stock using RSI. I knew that did not make sense. I ask him, did you use open with one day delay? He said no he used no delay. Well think what that does: it lets you buy at open price after the signal is given later that day. Then he had less than two years data he was looking at. I told him to change to 5 years and buy at next day's open after a signal, I bet his profit is cut in 1/2. You have got to understand system test limitations. It is not just, did it beat buy and hold or a great % profit? everyone should learn to read System test, MSWIN gives the best informational results in their tests.



To: Jan Robert Wolansky who wrote (7463)1/3/1998 7:30:00 AM
From: David R. Evans  Read Replies (3) | Respond to of 12039
 
Hello Jan,

If you test with 80/80 stocks then you always fill your database with 80/80 stocks. Why would you care about anything else?????

Stocks do not always STAY 80/80 stocks so you need to constantly review your database to ensure you have the best. It's not always easy BUT it's worth it if your making money.

Dave Evans