trendscanner
265 posts
msg #80992
- Ignore trendscanner
modified |
10/12/2009 9:08:07 PM
Stocks will sometimes bounce off a lower BB rather than penetrate it. So a close near the lower BB may provide a relatively safe entry point for a long position, particularly if that level coincides with some other support level, such as a previous support or a significant MA.
I’ve been working with a filter to try to isolate best conditions for taking advantage of this set up, looking for something that provides a consistent and profitable approach. Haven’t found the ideal combination of conditions yet but thought I’d post this as work-in-progress. Comments/suggestions for improvement are welcome.
I started with a code line that selects day ranger stocks, since I want something that has the potential to move significantly upward, then I picked a price and volume range that I like. Since I can’t code in a universal setting for past support for all stocks, I used the ema(100) as something that might provide additional support to the lower BB. Added some lines to select for a close near the lower BB. Through trial and error I found that having a slope of the close coming into the set up relatively flat to positive worked best. Also, T&E found that ensuring that the low of the day didn’t plunge too far below lower BB helped eliminate losers.
The current filter form is:
Backtest results for 4/1/09 through today, selecting for stocks within 5 % of EMA100, and using default exit criteria are:
Trade Statistics
There were 19 total stocks entered. Of those, 18 or 94.74% were complete and 1 or 5.26% were open.
Of the 18 completed trades, 16 trades or 88.89%resulted in a net gain.
Your average net change for completed trades was: 7.28%.
The average draw down of your approach was: -4.99%.
The average max profit of your approach was: 12.88%
The Reward/Risk ratio for this approach is: 7.55
Annualized Return on Investment (ROI): 271.82%, the ROI of ^SPX was: 65.64%.
Exit Statistics
Stop Loss was triggered 2 times or 11.11% of the time.
Stop Profit was triggered 15 times or 83.33% of the time.
Trailing Stop Loss was triggered 0 times or 0.00% of the time.
You held for the maximum period of time (25 days) 1 times or 5.56% of the time.
An exit trigger was executed 0 times or 0.00% of the time.
This doesn’t produce a lot of hits, maybe 2 to 3 a month on average, but they are consistently profitable for the time period tested. The average 2 day gain was 2.4%, the average 5 day gain was 7.5% and average 10 day gain was 11.8%.
Using a criteria of being within 8% of the ema(100) gave slightly better overall results and a few more trades, although the 2 day, 5 day, and 10 day results were a little better with the 5% criteria. For the 8% setting, the results are as follows. The Reward/Risk of 10.6 is very good, and confirms that the entry point identified with the filter is relatively safe.
Trade Statistics
There were 25 total stocks entered. Of those, 24 or 96.00% were complete and 1 or 4.00% were open.
Of the 24 completed trades, 22 trades or 91.67%resulted in a net gain.
Your average net change for completed trades was: 7.97%.
The average draw down of your approach was: -4.60%.
The average max profit of your approach was: 12.66%
The Reward/Risk ratio for this approach is: 10.56
Annualized Return on Investment (ROI): 314.93%, the ROI of ^SPX was: 65.64%.
Exit Statistics
Stop Loss was triggered 2 times or 8.33% of the time.
Stop Profit was triggered 21 times or 87.50% of the time.
Trailing Stop Loss was triggered 0 times or 0.00% of the time.
You held for the maximum period of time (25 days) 1 times or 4.17% of the time.
An exit trigger was executed 0 times or 0.00% of the time.
One note – this filter didn’t perform particularly well when backtested against previous years such as 2005 through 2008. This filter appears suitable only for this market. Not sure why. Hopefully some additional testing and creativity may find something that works a little more universally.
GLTA trading this week.
|