Optimizing EAs in Real Time - page 3

 
Marco vd Heijden:

I can only hear you yelling show me this, show me that, show me.....

I see youve decided to stop.

Best decission of today.

Thank you, But I never yelled, it is your mind that creating your perception of the yelling which only means that you were emotional not me. Sorry.

I have stopped not because I am wrong, because wasting time on thinking that anything might change, is a dream.

 
Carl Schreiber:

I don't understand your position(s), PCWalker?

1) The ZorroTrader is free for some basic strategies but right after that ...?

2) The access to R is available too in Mql4 or Mql5 - but as long as the user don't understand what the various options of R mean, what they can do (and what not) and how they can be used this access of of absolutely of no help. E.g. I guess that most of the EA coders do not know the meaning of the skewness and kurtosis and why they are relevant in some cases when you build an average and in some cases not.

3) Optimizing in real time sounds to me rather strange. Do you mean to run an optimization right after each trades that closes with a loss? I have seen system that just reverse the usage of the signals after a loss - but they were not successful at all!
Would that be the outcome more or less? I think if you want to achieve this you haven't understand that trading by a robot or a human is a kind of gamble. Would it make sense to change the gambling-system each time the ball has landed on a number. I guess you will lose a lot of money in a very short time and - if you reverse this - you wouldn't gain any money as well.

No you run the optimization process in relation to how much the market has changed relative to the last time you made the optimization process.
Reason: