CLICKABLE FILTERS - StockFetcher.com Stock Screener

StockFetcher Forums · Filter Exchange · CLICKABLE FILTERS

<< 1 2 3 4 >>Post Follow-up

corsino
259 posts
msg #35858
- Ignore corsino

5/2/2005 7:08:01 PM

leaddog
I had to try it a couple of times, but it seemed to work the second time. The way I seemed to remember was that some was done with the right or left mouse buttons. Your way is simpler. Thanks.

corsino
259 posts
msg #35862
- Ignore corsino

5/2/2005 8:15:42 PM

Cjust
The filter you posted is very respectable. It has a good combination of number of stock picks / ROI. I have found that the higher the ROI,the fewer the stock picks. Maybe one needs an arsenal of several high-ROI filters if they want to stay continously "invested" in swing trading. My highest-ROI filters have had somewhere between 14 and 35 picks so far this year,although a couple + 360% ROI filters have had about 74 picks.

glider
59 posts
msg #35866
- Ignore glider

5/3/2005 2:30:44 AM

Corsino,
Good work on the filters. Here's my 2 cents worth. Sometimes the results are not consistent.
Example:
M 600 - COG>1 + 10 % DECREASE.
Global Filter Off
No Stop Loss
No Stop Profit
5 Day Hold
---------------------------
Dates:12/31/04 - 4/29/05
ROI: 667.16%
---------------------------
Dates: 1/2/04 - 12/29/04
ROI: -77.06%
---------------------------
Why would there be such a difference in ROI?

corsino
259 posts
msg #35868
- Ignore corsino

5/3/2005 9:59:26 AM

Hi glider
There are things about this backtesting feature I don't understand. I tried to run M600 for the period you specify, but couldn't. I got the message "Test period cannot exceed 1 year", although the period is not more than a year.So I clicked on " Use MAx".The dates 1/2/04 to 7/2/04 appeared. So I ran that. The results:

win/loss: 64/27
ROI: 195.67 %

Maybe I should have run the last 6 months of 2004, but didn't think of it until now.

corsino
259 posts
msg #35873
- Ignore corsino

5/3/2005 10:29:48 AM

Glider
I've been pretty stupid in running my filters.After reading your post it dawned on me that my Global Filter was on. It was "Price between 0.5 and 4 and average volume above 100000",which is a slightly different price range (0.5 to 3) than my filters specify.Undoubtedly, that probably accounts for different results. Also, market conditions will have an effect.
As you see, I'm still learning this backtesting stuff.

glider
59 posts
msg #35874
- Ignore glider

5/3/2005 11:53:07 AM

Corsino,

Regarding "Test period cannot exceed 1 year", make sure the year is correct. I got that message a couple of times and that was always the problem. When I corrected it, everything worked ok. Even though 'use max' gives you about six months, when you manually enter one year, it works.
You're correct about 'Global Off'. I always select it.
Also be careful after saving a backtest strategy based on a saved filter. I've changed some values in the saved filter after setting the backtesting up and found it wasn't referencing the new values. I had to manually change the values in the backtesting section also.
One other thing regarding general results. Even though the market has been down so far this year, the ROI on your filters is impressive. Most filters don't stand the test of a bear market. But I still don't understand why it would return very poor results for 2004.
Keep up the good work.

corsino
259 posts
msg #35879
- Ignore corsino

5/3/2005 6:30:39 PM

Glider
The thought has occurred to me that "bottom fishing" filters may do better in a bear market than in a good market. So along that line, it may be that "momentum" filters may do better in a bull market. Of course, that is just speculation on my part.
Thanks for your tips about backtesting.

corsino
259 posts
msg #35882
- Ignore corsino

5/4/2005 1:43:57 AM

A CONSISTENT FILTER

This is the most consistent filter I've found for the test period of 12/31/2003 to the present.I'll just post the 2004 results, but it has done equally well so far this year.

M112

Show stocks where volume is above 100000
and close is near bottom linear regression line (45,1.0)
and rsi(2)is less than 1
and close dropped more than 10% in the last 1 day
and price is between 0.5 and 3

Approach Information

Approach Name: M112- RSI / 10% DROP / LRL

Test started on 12/31/2003 ended on 12/31/2004, covering 252 days

Filter used:

M112- RSI / 10% DROP / LRL (saved filter)

Trade Statistics

There were 120 total stocks entered. Of those, 119 or 99.17% were complete and 1 or 0.83% were open.

Of the 119 completed trades, 71 trades or 59.66%resulted in a net gain.

Your average net change for completed trades was: 5.22%.

The average draw down of your approach was: -10.46%.

The average max profit of your approach was: 19.33%

The Risk/Reward ratio for this approach is: 2.28

Annualized Return on Investment (ROI): 257.98%, the ROI of ^IXIC was: 8.03%.

Exit Statistics

Stop Loss was triggered 0 times or 0.00% of the time.

Stop Profit was triggered 0 times or 0.00% of the time.

Trailing Stop Loss was triggered 0 times or 0.00% of the time.

You held for the maximum period of time (5 days) 119 times or 100.00% of the time.

An exit trigger was executed 0 times or 0.00% of the time.

Statistics By Holding Period
	Completed	2 day chg	5 day chg	10 day chg	25 day chg	40 day chg
Winners:	71	81	71	60	56	56
Losers:	46	34	47	57	64	62
Win/Loss Ratio:	1.54:1	2.38:1	1.51:1	1.05:1	0.88:1	0.90:1
Net Change:	5.22%	5.28%	4.88%	1.40%	-0.45%	0.20%

Statistics By Variable: Match Price
	<0.6	<0.9	<1.2	<1.5	<1.8	<2.1	<2.4	<2.7	<3	<3.3
Completed	7:3	16:6	12:5	10:6	5:9	5:4	7:8	8:5	1:0	-
2 day chg	10:0	15:4	13:5	11:5	6:7	7:2	7:8	11:3	1:0	-
5 day chg	7:3	16:6	13:5	10:6	5:9	4:4	7:8	8:6	1:0	-
10 day chg	7:4	12:9	8:10	9:7	5:9	5:4	5:8	9:5	0:1	-
25 day chg	7:4	12:10	9:9	8:8	4:10	4:5	7:8	5:9	0:1	-
40 day chg	6:4	12:10	7:10	9:7	6:8	5:4	8:7	3:11	0:1	-

Statistics By Variable: Average Volume
	<1.5M	<3.0M	<4.5M	<6.0M	<7.5M	<9.0M	<10.5M	<12.0M	<13.5M	<15.0M
Completed	65:44	4:0	2:1	-	-	-	-	-	-	0:1
2 day chg	75:32	4:0	2:1	-	-	-	-	-	-	0:1
5 day chg	65:45	4:0	2:1	-	-	-	-	-	-	0:1
10 day chg	58:52	2:2	0:2	-	-	-	-	-	-	0:1
25 day chg	53:59	1:3	2:1	-	-	-	-	-	-	0:1
40 day chg	55:55	0:4	1:2	-	-	-	-	-	-	0:1

corsino
259 posts
msg #35883
- Ignore corsino

5/4/2005 2:04:23 AM

The M112 filter results were for a 5-day holding period with no stops. The results for a 2-day hold were better:

12/31/2003 to 12/31/2004....Win/loss :67/29 ROI: 714.95%

12/31/2004 to 5/3/2005.....Win/loss :58/41 ROI : 461.39%

Koronbock
201 posts
msg #35884
- Ignore Koronbock

5/4/2005 9:34:04 AM

Corsino,
thanks for posting your filter. It does indeed give good results. I do not want to spoil the party, just caution you before you go out and buy that flashy BMW:

REAL TRADING is VERY different from a backtest. I found this out the hard way. Volatility and bid/ask spread play a big role here, because you are dealing with very low priced stocks. Unless there is big volume (minimum 500 k)the bid/ask spread can easily be a few percent. Try your filter with a 500 k volume and you will see what I mean. Also when exiting you will almost certainly NOT get the Opening price as you must sell "at the market".

No question, your filter is very good. But I'd be surprised if you can consistently get even half of the ROI the backtest is showing. But again, this would be a VERY nice return still.

Cheers and good trading.

StockFetcher Forums · Filter Exchange · CLICKABLE FILTERS

<< 1 2 3 4 >>Post Follow-up