From theory to practice - page 75

New comment
 
Nikolay Demko:

Half a year is enough, you can learn MQL in a fortnight. And for those who know C, C++, C# and only a part of it is needed, they can learn it in a day.

Although I agree that in order to write an EA you need some experience, especially as a trader in order to understand all the details.


The main thing is practice! I also learned the language in a fortnight (syntax, loops, functions).

 
Vladimir:

I reread it, it really came out crooked. It's as if I'm trying to emphasise what I've got and you haven't got. I'm sorry if you thought the same. I didn't have that in mind, I really want to know how this kind of auto modeling can help.

I myself only use the square root law to narrow down the search, to find out "where to look". For example, the places where the theoretical maximum possible profit is greatest (where one spread-adjusted profit comes out per trade). This often allows you to practically reduce the dimensionality of the search space.

Never mind. I didn't post the results, just the conclusion. I didn't even bother saving the calculations, as I was interested in the presence or absence of it. To this day I do not know what it may be used for. Something did not see, what is your conclusion?

As for the experiment itself, they are almost equivalent. Only instead of MA I used LF filters, and instead of a "frequency" distribution (as far as I understood you applied it) I used a probability distribution. I reduced them together by converting the cutoff frequencies of the filters to unity, with a corresponding change in the amplitude by signal energy.

Alternatively, the same could be shown via the Fourier transform and comparison of spectra via scaling. This was not done.

 
Alexander_K2:
Let me explain again. The probability distribution you see on the increments - it doesn't go anywhere. It wants to live! It appears in one way or another in deviations of prices from the moving average. And the goal of these experiments that Vladimir has done is to understand at which MA the deviations from it form a distribution that is maximally similar to the distribution that you see on the increments. That will be the sampling volume for calculation of a trailing MA that will maximize a profit. What's not to understand?

Why did you decide that your distribution on the increments should be the same as the distribution on the deviations from ma.

 
Alexander_K2:
Let me explain again. The probability distribution you see in the increments - it doesn't go anywhere. It wants to live! It shows up in one way or another in price deviations from the moving average. And the goal of these experiments that Vladimir has done is to understand at which MA the deviations from it form a distribution that is maximally similar to the distribution that you see on the increments. That will be the sampling volume for calculation of a trailing MA that will maximize a profit. What's not to understand?
That's not what we mean.)
 
Alexander_K2:

And for nothing!

Let me explain. I was answering Vladimir's specific question.
 
Alexander_K2:

:))))))))))))))) That's just the way I am. As someone here said, "I demand a continuation of the banquet!" :))))))


What banquet? Show me your intermediate results! )

 
Максим Дмитриев:

What banquet? Show me your intermediate results! )

And from the audience they shout to me - Give me the details! (с)


Oh, you'll have to write an article, Alexander.) By the way, they give money for articles.) The real ones, not the demo ones.

 
Alexander_K2:
Um... No, I'm lazy. I'm used to exclusively supervise, but if some student, from those present here, will write a term paper on stationarity/non-stationarity on the basis of nonparametric methods, and will post it here with a supervisor's signature - that'll be YES!

I was inspired by it. And as an old and experienced provocateur, I couldn't resist).

I'm not a scale or half a scale, I sat chewing on a black bread, I kept looking at Ksen'kin's bespectacled boy, the way he pretended to be a scientist. 

Maybe I'm a seminarian myself. 
Maybe I'm a chauffeur for the occasion, I see even the guests have fallen asleep, Even Ksen'ka, I see, is a cloud in a cloud. 

But he sings like a choir at all-night, All about Ixes, Ixes and sines, And the suit is nothing to see: 
A tailor's tailor's tailor's tailor's tail. 

And he doesn't live in Dubna, but in some research institute near Kashira. 

He'll tell her, "Multiply twenty-five by nine and one hundredth." And then he'll sit there, kicking and moaning, while she works. 

And she works without a murmur, The lights on the remote control are streamlined! 
Well, what are we, among the robots, What are we to do, the people who are lost?!
 
Alexander_K2:
... If any student here writes a term paper...

Students are extremely rare here. There are very few young people (under 30) either. The main contingent is between 40+ and 70+, so it's a mistake to be old. Your age status on this forum is closer to that of the young.

 
Alexander_K2:

Hmmm... What's taking you gentlemen so long with this task???


1...686970717273747576777879808182...1981
New comment