Retraining - page 6

New comment
 
Yousufkhodja Sultonov:
What is your problem with this kind of "fit"? For 40 years we have got MO = 6.02 points/day, profitability = 2.06. For the last 3+ years, MO = 7.26 and Pr = 2.66. The figures are of the same order of magnitude.
Just do awolfing forward of your EA and then there will be something to talk about. Including in relation to the topic of the topic of the topic.
 
Youri Tarshecki:
The misunderstanding is that you're not saying - I've had these settings for 40 years and lo and behold, they work on the last unoptimised three as well. You are talking about the same settings. But it is not known how you got them. Judging by your products and posts - you just optimised the whole story. And this is the fitting. Just do a wolfing forward of your EA at least in the mode of 10 year optimization - 5 year check and then you yourself will see the sad result.
Who told you that the"volking forward of your EA at least in 10-let-optimization-5-year check mode" is better than I do? What is the advantage of your method? Who forbids me to do as I have shown? Why should I crush the available history and not use the whole? Do you have research results on both options? What exactly 10 and 5 years should I take? Why do I need such uncertainty and conundrum? Isn't it more logical to assume that if TC has passed the whole story, then, there is a high probability of passing it in the future as well?
 
Yousufkhodja Sultonov:
Who told you that, the option of"volking forward your EA at least in the 10 year-optimisation-5 year check mode" is better than what I do? What is the advantage of your method. Who forbids me to do it the way I showed you?
The advantage is that this is the only way to simulate for your EA the situation of non-optimized future. If you are so confident about your EA - doa wolfing forward and show that I'm wrong. And then we can discuss whether your EA is overoptimized or not in these areas. When you say that you have optimised your EA on the whole story and the future will be exactly the same - this is cheating, because I am being asked to wait another 43 years to check whether you are right or not.
 
Youri Tarshecki:
Just doa volking forward of your EA and then there will be something to talk about. Including in relation to the topic of the topic of the topic of the topstarter.

From 2000 to 2010:

From 2010 to 2016 (present):


 
Youri Tarshecki:
The advantage is that this is the only way to simulate an unoptimised future situation for your EA. If you are so confident in your EA - do awolfing forward and show that I'm wrong. And then we can discuss whether your EA is overoptimised or not on the given plots. When you say that you have optimised your EA on the whole story and the future will be exactly the same - this is cheating, because I am invited to wait another 43 years to check whether you are right or not.
Why are you so insistent that "this is the only way to simulate an unoptimised future situation for your EA"? What is the element of "fraud"? Forbidden to use the whole story? Who forbids me? I have to admit, the efficiency of the TS is low, but I don't expect more from the market.
 
Yousufkhodja Sultonov:
Why are you so insistent that "this is the only way to simulate an unoptimised future situation for your advisor"?
Enough demagoguery. Just do a wolfing forward and show it. And when it shows the sadness, let's discuss what's over-optimised and what's not.
 
Youri Tarshecki:
Enough with the demagogy. Just do a volking forward and show it. And when it shows the sadness, let's discuss what is over-optimised and what is not.

I.e., I should optimize TC from 2000 to 2010 and with these settings go through data from 2010 to 2016?

I did this operation and it turned out that the optimization parameters (N-sample volume, TP and SL) were completely the same as the optimization results for 43 years. I think repeating the passes is pointless.

 
Yousufkhodja Sultonov:
I.e. I should optimize my TS from 2000 to 2010 and with these settings go through the data from 2010 to 2016?

No, we start with, say, 1975.

Optimisation 1975-1985, verification 1985-1990

Optimisation 1980-1990, verification 1990-1995

Optimisation 1985-1995, check 1995-2000

Optimization 2000-2005, verification 2005-2010

Optimization 2005-2010, verification 2010-2015

Look only at test results, and if at least ONE of these five years will be negative (and I think there will be more), then the system is defective.

I.e. your trick with fitting on the whole history will only work if there are crazy people willing to wait for decades for profits from your EA.

And by the way, don't forget to tell me how you avoid over-optimization on each plot.)

 
Youri Tarshecki:

No, we start with, say, 1975.

Optimisation 1975-1985, verification 1985-1990

Optimisation 1980-1990, verification 1990-1995

Optimisation 1985-1995, check 1995-2000

Optimization 2000-2005, verification 2005-2010

Optimization 2005-2010, verification 2010-2015

We look only at test results, and if at least ONE of these five years will be negative (and I think there will be more), then the system is defective.

That's assuming there are crazy people who are ready to wait for decades for profit from your EA.

And by the way, don't forget to tell us how you avoid over-optimisation in each area.)

I will, although I don't believe in the correctness of this approach, apparently this erroneous postulate sits deep in your subconscious. Explain clearly what pre- and/or over-optimisation means. I do not understand the invention of super-traders.
 
Yousufkhodja Sultonov:
I will do it, although I do not believe in the correctness of this approach, apparently this erroneous postulate is sitting deep in your subconscious.
Do it, but don't cheat. I would do it myself, but my auto-optimizer is set to MT5, and I don't feel like redoing it.
1234567
New comment