DAX 1 min active algo

• 09/12/2019 at 3:24 PM #107356
  Stefanb

  Well, what do you think?

  Optimized the code on 70% data.

  Made a WF of 5 periods of 70%.

  Here is the result.

  Do you think the code looks good or did I miss something?

  Was looking for an active algo, this one makes 3 trades a day.

  The reason I want an active algo is because I am tempted to do manual trades if my algo is flat.

  Then it really goes to hell

  It’s a simple code that you see.

  Have heard a lot of people say that of all the codes you have, it is the easiest one that goes best, you should perhaps not complicate it?

  DAX 1 min

  Spread=1

  

  Stefanb

  Participant

  Topics: 15

  Replies: 74

  Been thanked: 22 times
  09/12/2019 at 3:26 PM #107358
  Vonasi

  What code? You haven’t shared it. 🙂

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
  09/12/2019 at 3:40 PM #107359
  Stefanb

  Sorry for that..

  The code.

  // Definition of code parameters DEFPARAM CumulateOrders = False // Cumulating positions deactivated TIMEFRAME (15 minute, default) // Conditions to enter long positions indicator1 = Average[53](close) c1 = (close CROSSES OVER indicator1) indicator3 = RSI[14](close) c3 = (indicator3 > 10) // Conditions to exit long positions indicator2 = Average[35](close) c2 = (close CROSSES UNDER indicator2) // TIMEFRAME (default) IF c1 AND c3 and Not OnMarket THEN BUY 1 CONTRACT AT MARKET ENDIF IF c2 AND LongOnMarket THEN SELL AT MARKET ENDIF // Stops and targets // --- NOT ALLOWED ---> SET STOP pLOSS a pTRAILING b //uncomment one of the following lines //SET STOP pLOSS a SET STOP pTRAILING 59

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

  // Definition of code parameters DEFPARAM CumulateOrders = False // Cumulating positions deactivated   TIMEFRAME (15 minute, default) // Conditions to enter long positions indicator1 = Average[53](close) c1 = (close CROSSES OVER indicator1) indicator3 = RSI[14](close) c3 = (indicator3 > 10) // Conditions to exit long positions indicator2 = Average[35](close) c2 = (close CROSSES UNDER indicator2) // TIMEFRAME (default) IF c1 AND c3 and Not OnMarket THEN BUY 1 CONTRACT AT MARKET ENDIF IF c2 AND LongOnMarket THEN SELL AT MARKET ENDIF // Stops and targets // --- NOT ALLOWED --->  SET STOP pLOSS a pTRAILING b //uncomment one of the following lines //SET STOP pLOSS a SET STOP pTRAILING 59

   

  

  Stefanb

  Participant

  Topics: 15

  Replies: 74

  Been thanked: 22 times
  09/12/2019 at 4:54 PM #107368
  Vonasi

  Whenever I see something with averages as entry and exit filters I worry about curve fitting and so like to optimise each one separately and see how important each variables variable is by graphing them in Excel. Attached is the image for the period in the exit average for values 2 to 52. Clearly there is a peak curve fitted value for the data sample you have – either side profit and performance falls away rapidly. Perhaps a little curve fitted.

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
  09/12/2019 at 5:37 PM #107372
  Stefanb

  @Vonasi

   

  Thanks for your input.

  You have a point there.

  Thought to make a simple algo, but it was perhaps too simple.

  Let’s add some filters and see if the result stands  🙂

  

  Stefanb

  Participant

  Topics: 15

  Replies: 74

  Been thanked: 22 times
  09/12/2019 at 10:58 PM #107396
  Vonasi

  My quick test of one variable does not mean that as a strategy it is a complete failure it just says that that one variable is quite optimised for the data we are testing it on. On a 1 minute time frame we have very limited data for in sample and out of sample testing. You have found something that works on the sample we have. Perhaps if we re-optimised the variables very often it would keep working… or perhaps we would always be just slightly behind the curve on such a fast moving time frame?

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
• Author
  Posts