Long only 15 min DAX
-
Hi
I just started working on a system which is buying pullbacks on Dax 15 min. I would be grateful if anyone had the possibility to test the system on a longer history. The % of winning trades are not very impressive so if anyone have suggestions on how to improve this I would be grateful to hear them. I’m a beginner when it comes to this so any feedback on how to optimise the code is appreciated!
(The backtest I’ve done is with 1p spread)
cheers
// Definition of code parameters DEFPARAM CumulateOrders = False // Cumulating positions deactivated // Prevents the system from placing new orders on specified days of the week daysForbiddenEntry = OpenDayOfWeek = 6 OR OpenDayOfWeek = 0 // Conditions to enter long positions indicator1 = TEMA[10](close) indicator2 = SuperTrend[3,10] c1 = (indicator1 >= indicator2) indicator3 = CCI[20] c2 = (indicator3 CROSSES OVER -150) IF (c1 AND c2) AND not daysForbiddenEntry THEN BUY 1 CONTRACT AT MARKET ENDIF // Conditions to exit long positions indicator4 = CCI[20] c3 = (indicator4 CROSSES UNDER 100) IF c3 THEN SELL AT MARKET ENDIF // Stops and targets SET STOP pLOSS 25Paris thanked this postHi – Thanks for sharing your code. I have tested it on a longer history going back to 2009 and the good news is that it still wins. However the gains are small compared to the period and no of trades conducted and more importantly it only really starts performing from Oct 2015 onwards (where it has some quite nice results).
I have attached the results (Test 1) together with amending the stop loss to 0.9% (Test 2) as a % based stop loss provides better results given that the DAX was half its current value back then. It produces better results but still needs some improvement. The draw is still only 7% which is pretty good. I’ll have a think about other ways you can improve. Have you optimised anything yet ? It doesn’t look like it, so together with the low draw of 3-7% you’re off to a good start.
Paris thanked this postI have now added a short condition similar to the long condition I’m using. It makes up for when the market is more bearish. The best return 2 years history I have is given by a 0,8% stop. Any ideas for improvements are most welcome!
// Definition of code parameters DEFPARAM CumulateOrders = False // Cumulating positions deactivated // Prevents the system from placing new orders on specified days of the week daysForbiddenEntry = OpenDayOfWeek = 6 OR OpenDayOfWeek = 0 // Conditions to enter long positions indicator1 = TEMA[10](close) indicator2 = SuperTrend[3,10] c1 = (indicator1 >= indicator2) indicator3 = CCI[20] c2 = (indicator3 CROSSES OVER -100) IF (c1 AND c2) AND not daysForbiddenEntry THEN BUY 1 CONTRACT AT MARKET ENDIF // Conditions to exit long positions indicator4 = CCI[20] c3 = (indicator4 CROSSES UNDER 100) IF c3 THEN SELL AT MARKET ENDIF // Conditions to enter short positions indicator5 = TEMA[10](close) indicator6 = SuperTrend[3,10] c4 = (indicator5 < indicator6) indicator7 = CCI[20] c5 = (indicator7 CROSSES UNDER 200) IF (c4 AND c5) AND not daysForbiddenEntry THEN SELLSHORT 1 CONTRACT AT MARKET ENDIF // Conditions to exit short positions indicator8 = CCI[20] c6 = (indicator8 CROSSES OVER -200) IF c6 THEN EXITSHORT AT MARKET ENDIF // Stops and targets SET STOP %LOSS 0.8Paris thanked this postAbout filtering range, it has always been something not really simple. You can give a try with (for instance): Keltner channel inside/outside Bollinger Bands, or adapt this indicator’s code I posted recently in the library (Volatility breakout indicator), just 2 rough ideas this morning 🙂
Please find attached the backtest of the system in tick mode with 1.5 points spread over 200.000 units. Symmetry of trading conditions for long and short orders might not be a good option if you plan to stick on a 15 minutes timeframe. The long term trend seems to affect a lot this concept.
// Definition of code parameters DEFPARAM CumulateOrders = False // Cumulating positions deactivated // Prevents the system from placing new orders on specified days of the week daysForbiddenEntry = OpenDayOfWeek = 6 OR OpenDayOfWeek = 0 // Conditions to enter long positions indicator1 = TEMA[10](close) indicator2 = SuperTrend[3,10] c1 = (indicator1 >= indicator2) indicator3 = CCI[20] c2 = (indicator3 CROSSES OVER -100) indicator4 = SuperTrend[3,10] indicator5 = Average[2200](close) c3 = (indicator4 > indicator5) IF (c1 AND c2 AND c3) AND not daysForbiddenEntry THEN BUY 1 CONTRACT AT MARKET ENDIF // Conditions to exit long positions indicator6 = CCI[20] c4 = (indicator6 CROSSES UNDER 100) IF c4 THEN SELL AT MARKET ENDIF // Conditions to enter short positions indicator7 = TEMA[10](close) indicator8 = SuperTrend[3,10] c5 = (indicator7 < indicator8) indicator9 = CCI[20] c6 = (indicator9 CROSSES UNDER 200) indicator10 = TEMA[10](close) indicator11 = Average[2200](close) c7 = (indicator10 < indicator11) IF (c5 AND c6 AND c7) AND not daysForbiddenEntry THEN SELLSHORT 1 CONTRACT AT MARKET ENDIF // Conditions to exit short positions indicator12 = CCI[20] c8 = (indicator12 CROSSES OVER -200) IF c8 THEN EXITSHORT AT MARKET ENDIF // Stops and targets SET STOP %LOSS 0.8Paris thanked this post
- You must be logged in to reply to this topic.
Author
Summary
This topic contains 27 replies,
has 5 voices, and was last updated by
8 years, 11 months ago.
Topic Details
| Forum: | ProOrder: Automated Strategies & Backtesting |
| Language: | English |
| Started: | 03/15/2017 |
| Status: | Active |
| Attachments: | No files |
