HOW LONG PROBACKTEST PERIOD?!

• 04/28/2017 at 10:42 AM #33824
  ALE

  Markets change, how do you choose the historical period to test a strategy?

  

  ALE

  Moderator

  Topics: 17

  Replies: 570

  Been thanked: 223 times
  04/28/2017 at 10:56 AM #33828
  Maz

  General rule of thumb:

  The higher the time-frame, the longer the back test. Depends on how long you expect to run the algo. What would be nice is to do some benchmark testing and get some standards, ie:

  • >= H4 :: 10 years
  • >= M15 < H4 :: 5 years
  • <M15 :: 2 years

  You could argue that one complete boom-bust cycle would be prudent. I think that idea is good for H1 and above, especially above H4 systems. But because technology moves on at an ever increasing pace, trading systems themselves as well as the instruments available at the beginning of the last boom-bust cycle are so different to what we have now. For the smaller time frames (particularly market micro-structure, S1 and below) you are looking at different kinds of price action dynamic shits than larger time frames).

   

  

  Maz

  Participant

  Topics: 14

  Replies: 121

  Been thanked: 139 times
  04/28/2017 at 10:58 AM #33829
  Maz

  Same goes for non temporal systems

  

  Maz

  Participant

  Topics: 14

  Replies: 121

  Been thanked: 139 times
  04/28/2017 at 11:05 AM #33832
  Maz

  …and ideally you want to capture huge macroeconomic events or flash crashes to stress test your system. Referendums; presidential elections; central bank announcements; interest rate changes; macroeconomic announcements and disasters – they mess up highly technical systems; especially one that run on higher time frames which is why you need to go back at least enough to capture performance during these events.

  Also note that backtests don’t simulate slippage; execution errors; liquidity problems; shifts in spread; market closures and so on. A better backtest framework would incorporate simulated slippage during high volatility, changes of spread, the odd trade rejection etc etc!

   

  

  Maz

  Participant

  Topics: 14

  Replies: 121

  Been thanked: 139 times
  04/28/2017 at 12:31 PM #33844
  ALE

  Thanks Maz,
  So the rule that says “Strategy must have at least one 10-year history” is not valid for all time frames, and it also seems to be too static if one takes into account the technological changes in place. Can we consider sleepage and macroeconomic data/events  neutral event? I mean we have 50% of probability, it’s correct?

  

  ALE

  Moderator

  Topics: 17

  Replies: 570

  Been thanked: 223 times
• Author
  Posts