DaveN please

Discussion in 'Trading' started by jrkob, May 27, 2004.

  1. jrkob


    DaveN, I am refering to the following thread you launched a long while ago:


    I have 2 questions:

    1) I do not understand "This optimization consisted of 20 steps, so I'd take (1-.10)=90% chance of success in the future or out of sample data raised to the power of 20, (.90^20)=11% !!! So much for a good chance of profits in the future! This is a great example of overoptimization."

    I have a background in probability, however I don't understand why you raise to the power of 20 ?

    2) See the attached file. They are hypothetical net PnL of trades resulting from an hypothetical trading system, in the order they appeared. As you can see, the system generated 1069 trades. My t-test calculation show that my confidence level is close do 100%.
    Is that correct ? If yes, it merely means that my sample is large enough to be reliable, is that correct ?

    However, it doesn't address the problem of the “sequence” in which the profits and losses appear. If the profits were appearing first, then followed by all the losses, I would certainly not trade that system, even though the t calculation remained the same.

    In other words, the t-test only address the question of whether we have backtested enough to be statistically significant.

    Is that correct ?

    Thanks in advance.
  2. jrkob


    DaveN or anybody who has the answers to my questions of course !