shelupinin
120 posts
msg #36342
- Ignore shelupinin |
6/15/2005 4:08:15 AM
this filter shows stock which are bouncing , RSI(2)<10 filters stock which at the bottom, so there is high probability that it will bounce up
Try to backtest this filter with 1% stop loss, without any profit stop and maximum holding perion 1 day. Depending of the time frame of test this filter will show ROI from 400% to 1000%, for example for backtest period from 06/03/2004 to 06/03/2005 the ROI=1052.26% ! Not bad result for filter which uses marginable stocks over 5 USD !!!
|
MACDONALD
18 posts
msg #97781
- Ignore MACDONALD |
12/3/2010 2:09:35 PM
I am new to this, and this is a really great filter! What sort of entry points, or buy triggers to you use for this?
Or do you just enter at opening at set the stop @ 1%..or ??
Thanks in advance
MacDonald
|
MACDONALD
18 posts
msg #97783
- Ignore MACDONALD |
12/3/2010 2:16:50 PM
Any addtional advise on buy triggers...to avoid head fakes.
Thanks
Macdonald
|
MACDONALD
18 posts
msg #98589
- Ignore MACDONALD |
1/15/2011 2:11:10 PM
I have tried making a "Shorting version" of this filter, but no success. Any suggestions on this anyone?
Thanks
|
jimvin
173 posts
msg #98662
- Ignore jimvin |
1/19/2011 11:18:04 PM
Every post deserves a reply...so here's my moderately informed resposne. I've been researching/using stock filters/models for a few years (made a lot of money/lost a lot of money) and one of my basic conclusions is that every market has its own personality and its own "play style."
Until about August of last year I had pretty good success (avg return of 5% = a week) playing a model that used % Advances, P&F Charts, MACD and classic RSI patterns to pick winners. In August, however, the market stopped playing by the rules I was using to win.
For example, one of the classic rules for using RSI is that stocks above 80 are overbought and are poison, while stocks below 20 are oversold and are fantastic bargains. Yet at about that point (August, 2010) stocks in the 80s and even 90s started going substantially higher in weekly plays...very much against classical theory and what had worked for me up until then.
My tentative conclusion is that this filter may well work (and may work well) but you should play it on paper for at least a couple of months tracking weekly or daily picks for a variety of time-frames (day, week, multiple weeks) and see what, if any, time-frame works.
(All of the above is my humble opinion based on humbling experience...)
|