Curve-fitting to success?

Discussion in 'Automated Trading' started by RStrauss, Sep 30, 2024.

  1. Handle123

    Handle123

    LOL, tried variations for couple of decades, never worked well.
     
    #11     Sep 30, 2024
    RStrauss likes this.
  2. RStrauss

    RStrauss

    what's variations?
     
    #12     Sep 30, 2024
  3. RStrauss

    RStrauss

    Thanks so much for your pointers! i use optimization to find the average measured move from a fixed point in time to another
     
    #13     Sep 30, 2024
  4. RStrauss

    RStrauss


    "You might have a luck for some months going forward after testing, but that is usually a much too short period and also it happens very rare that you have something working in the future from now."

    Yes that's also (sadly) my experience so far. i found it hard to trade discretionarily, to me it is no different from gambling. and there is too much emotion attached I often overtrade and end up regretting mightily.
     
    #14     Sep 30, 2024
  5. Hello RStrauss,

    There is only one way to know.

    There is only one way to know the answer to your question.

    Curve fit 100% and see what happens in the future.
     
    #15     Sep 30, 2024
    RStrauss likes this.
  6. RStrauss

    RStrauss

    Hey SML,

    I can tell you now. I curve fit for about a year and so far failing badly lol but thanks to position sizing and risk management I was able to break even. So confused now and thinking of giving up...

    Any good advice?

    Thanks.
     
    #16     Sep 30, 2024
    SimpleMeLike likes this.
  7. Hello RStrauss,

    Do not give up buddy.

    The only advice I can say is keep on trying. Keep on trying and do your best.

    That is all you can really do.
     
    #17     Sep 30, 2024
    RStrauss likes this.
  8. RStrauss

    RStrauss

    Sometimes I don't know if you really mean it or just trolling :)
     
    #18     Sep 30, 2024
    SimpleMeLike likes this.
  9. ph1l

    ph1l

    Marcos Lopez de Prado in "Advances in Financial Machine Learning"
    https://www.amazon.com/Advances-Financial-Machine-Learning-Marcos/dp/1119482089
    agrees about walk-forward optimization:
     
    #19     Sep 30, 2024
    RStrauss likes this.
  10. trismes

    trismes

    The problem is structural— all data driven models will be subject to curve fitting to a degree. If you’re not keen on the result, change the methodology
     
    #20     Sep 30, 2024
    RStrauss likes this.