backtesting is stupid

Discussion in 'Strategy Building' started by jonp, Jul 15, 2010.

  1. jonp

    jonp

    ok, well you could consider it backtesting then if that's your definition of it. i'm referring to real scientific analysis.

    admittedly my entry was a complete botch as of now, but i don't feel too bad or have stop losses below any so called support levels. i'm waiting this one out until tuesday morning before making a decision.
     
    #31     Jul 16, 2010
  2. Can you define this for us precisely?

     
    #32     Jul 16, 2010
  3. We are observing the Dunning-Kruger effect at work. Real-time training in recognizing it! Gotta love ET! Everyone but the OP gets an education!
     
    #33     Jul 16, 2010
  4. jonp

    jonp

    i'm done with this thread. white flag
     
    #34     Jul 16, 2010
  5. In parting, a little advice. Ever heard of Neuro-Linguistic Programming? If not, it's appropriate to learning about onself in relation to others. My favorite NLP quote is: "The meaning of your communication is the response you get."
     
    #35     Jul 16, 2010
  6. jonp

    jonp


    No I have not heard of it - I will look it up.
    Have a good weekend posters.
     
    #36     Jul 16, 2010
  7. Maybe you should add a few more words. What drove you to this despair, for example?
     
    #37     Jul 16, 2010
  8. My point was, if you're going base your trading decisions on conclusions that you make from historical data anyway, why not add some clarity and objectivity to the process by formally backtesting instead of winging it?

     
    #38     Jul 16, 2010
  9. jonp

    jonp

    i would rather wing it. not to get back into it but to be honest i'm still not entirely convinced that formal backtesting produces any edge or provides any form of clarity.
     
    #39     Jul 16, 2010
  10. alfobs

    alfobs

    IMHO, backtesting is the only way to get at least some confidence in a system before you start using it. So backtesting cannot be stupid for this simple reason. In fact a deep backtesting (with sufficient amount of data) can cover most of the market moves to prove a system on robustness, and therefore to take care of those cases in future. If somebody failed with trading after backtesting then it happened because of over fitting and lack of variety of market conditions in the historical data. Less indicators, some sound logic behind your algorithm (which must be no more complicated than should), implemented money and risk management will help to make backtesting valid. For instance my one year backtesting of 15 min bars have supported me for last 6 month without any significant deviation from backtest performance results. Also in my opinion forward testing does not have any advantage over properly organized backtesting process.
     
    #40     Jul 16, 2010