Machine learning in trading: theory, models, practice and algo-trading - page 2101
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Very interesting, thank you for sharing. As soon as I figure out my current tasks, I'll try to run the code.
)) No, of course not
go ahead, we'll have a laugh... but in detail
Maybe someday.
I have something to write, but because of the reluctance of developers to see the code for R in the articles - I will not.
prohibited?
better read carefully ))
Look, you have a problem - you need to find a function (a curve) from nothing,
I don't know what it should look like, I don't know what it depends on, the only thing I know is what I want to get out of it.
How will you solve such a problem?
I suggested a solution, to assemble this function from the sum of the harmonics.
Is that a trivial solution? I don't think so.
Wanted to share, I think it's interesting, and there are silly questions - commission, train-test.... This post is not about that at all, ALLO)
I did harmonic decomposition in the 1st or 2nd year of uni. So it's trivial. Millions of students have dabbled in it. And there were threads on the forum about spectral decomposition of quotations and there were articles.
Our task is not to look at pretty pictures, but to beat at least the spread with the commission. This is more important for me.
Hmmm, a thought occurred to me, is it possible to run the Python package for Metatrader5 in R-ka and use all the charms of Metatrader?
Well, that's what I do. All Python stuff works in R as native.
)) No, of course not.
maybe someday.
forbid?
Not willing :)
Decomposition into harmonics in the 1st or 2nd year of uni. So it's trivial. Millions of students dabbled in it. And there were topics on the forum about spectral decomposition of quotations and there were articles.
Our task is not to look at pretty pictures, but to beat at least the spread with the commission. This is more important for me.
Searching for an unknown function and Fourier series decomposition are NOT the same thing, even remotely.
That's what I do. Everything Python works in R as native.
Do you have any examples of how it works?
Man, it's a pity you don't write more, I've read some of your articles six times already...
So many topics still uncovered, so many interesting ....
Don't want to :)
faggots!
I have no other words.
Our goal is not to look at pretty pictures, but to beat at least the spread with the commission. That's more important to me.
So you should trade where the spread is small. For Si the spread is 1-3 pips, but the commission is at least 1.5 spreads.
So you should trade where the spread is small. In Si the spread is 1-3 pips, but the commission is 1.5 spreads.
I suggested to add only 0.00005 (very good spread + commission) - the curve will immediately go down.
There, 2000 trades give 0.04 of profit, i.e. 0.00002 per trade. There is no such a spread.
And I suggested to add only 0.00005 (a very good spread + commissions) - the curve will immediately go down.
There, 2,000 trades give 0.04 pips totaling 0.00002 per trade. There are no such spreads.
I think this is just an approach that can be developed until its potential is clear.
By the way, machine learning works better at the stock exchange - I'm comparing it with quotes from MQ - I cannot achieve stable growth at EURUSD after 1.5 years with 3000 if 1 lot, which is obviously not enough.
So write more. I'll even suggest a topic - machine learning automation. This is what almost all participants of this thread need.
Is the automation of MOA an automatic selection of the type of NS and MOA (optimized packages for BP) for optimal results?
Or just automation of learning process - easy preparation of data and getting ready TS?
And I suggested to add only 0.00005 (a very good spread + commissions) - the curve will immediately go down.
There, 2,000 trades give 0.04 pips totaling 0.00002 per trade. There are no such spreads.
Even by eye, from the picture with the trades you can see that the commission will not change anything ....
What kind of people are....