The "Silver Range M15" test

• 01/16/2017 at 9:44 AM #21237
  Doctrading

  Hello everyone,

  I wanted to code a breakout strategy on Silver.

  The results were completely opposite to those I hoped for : negative and losing regularly.

  So I had the idea of ​​reversing the parameters : instead of a breakout, we play a rebound on a “range” (including the last 22 candles), and moreover according to the opposite of the current trend (defined by moving averages 5 and 30).

  Of course, the profit factor is not optimal (1.13) ; and gains are not miraculous for the drawdown exposed (17%), according to the leverage used.

  But this strategy has the merit of being able to be deepened / optimized.
  

  And maybe there are mistakes with the test . I don’t have yet the 1.3 version or ProRealTime (which allows tick by tick backtest).

  // SILVER M15 DEFPARAM cumulateorders = false n = 4 SL = pipsize * 15 TP = pipsize * 30 MM5 = Average[5](close) MM30 = Average[30](close) // ACHAT IF MM5 > MM30 THEN sellshort n shares at highest[22](high) limit ENDIF IF MM5 < MM30 THEN buy n shares at lowest[22](low) limit ENDIF set stop loss SL set target profit TP

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

  // SILVER M15 DEFPARAM cumulateorders = false   n = 4   SL = pipsize * 15 TP = pipsize * 30 MM5 = Average[5](close) MM30 = Average[30](close)   // ACHAT   IF MM5 > MM30 THEN sellshort n shares at highest[22](high) limit ENDIF   IF MM5 < MM30 THEN buy n shares at lowest[22](low) limit ENDIF   set stop loss SL set target profit TP

  

  Doctrading

  Participant

  Topics: 44

  Replies: 172

  Been thanked: 2 times
  01/16/2017 at 10:15 AM #21245
  Nicolas

  Hi Doc, I moved your post from library pending review queue to forum.

  I agree that silver XAGUSD is a mean reverting instrument for intraday timeframe, you discover it also and it would be interesting to explore further more this concept with it.

  I think your tests were made without spread? Because I really don’t have the same results as yours.

  I have tested from 2011 May 1st, because it is the date where strategy begin to make money. Because tick per tick backtest is only available through demo account, I don’t have many allowed bars to use, so the backtest is stopped at the end of year 2012. You’ll find below screenshots of the tested strategy, with and without the spread (took 2.5 points only).

   

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4775 times
  01/16/2017 at 10:48 AM #21254
  Doctrading

  Hello Nicolas,

  You are right, I forgot spread… 🙁

   

  

  Doctrading

  Participant

  Topics: 44

  Replies: 172

  Been thanked: 2 times
• Author
  Posts