Not logged in
Viewing 2 posts - 1 through 2 (of 2 total)
  • #111665 |
    Customer
    30 Posts
     
    Hello!
     
    Guys, do you believe this method is infallible to determine the robustness of any EA? or do you think that there are exceptions due to the characteristics specific of any EA and possibly with some EA could skip this method?
     
    Thank you!
     
    AndrEAs

    AndrEAs

    #122846
    Mark Fric
    Administrator
    1182 Posts

    Hello,

     

    Monte Carlo method is an important tool, it tells you how your strategy behaves when input conditions are slightly changed.

    Ideally you want your EA to be robust and don’t break apart when parameters or data change slightly.

     

    This is what differences curve fitted strategies from robust ones – robust strategy should “survive” also Monte Carlo test.

     

    I wouldn’t skip Monte carlo test, maybe you can experiment with its setting and find what type of test (there are few different like skipping orders, randomizing history data, randomizing parameters, etc.)

    cause your strategy failure. Then you’ll know what your strategy is sensitive of.

    But I usually use the default setting and expect the strategy to pass the test completely.

     

    Mark

    Mark
    StrategyQuant architect

Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic.