Well, if you know how to work with this in other languages, then I think by analogy you can use this version as well. The I/O parameters should be similar
What does this Markov chain do anyway? In V. Perervenko's article it is described as a smoothing function and corrects the result of the NS.
Do you also want to flatten the series? Judging by the order of the record ("switching + ML") - before the training?
Holy crap, you guys... what the hell is smoothing? Who are you listening to... You're talking nonsense, ZZ in the targets, now it turns out hmm - smooths out the series))) Markovchain rather mediocre predicts sequences as probabilities of change states, you can certainly adapt, it was even on the site Fischmann Foundation (quantum brains capital), they wrote that they use Markovchain, about 6 years ago. Now they use LSTM.
I read the articles. One also uses it for correction, the other for trading. Something about Markov chains reminded me of clustering (also without a teacher and you can get several modes/clusters to choose from).
Read the articles. One also uses it for correction, the other for trading. Something about Markov chains reminded me of clustering (also without a teacher and you can get multiple modes/clusters to choose from).
Yes, then detect them in the future and choose a model that works well on this particular cluster
Allow the use of cookies to log in to the MQL5.com website.
Please enable the necessary setting in your browser, otherwise you will not be able to log in.
You are signed in. Two-factor authentication has been disabled. Set it up again and scan a new QR code, or continue with two-factor authentication disabled.
Well, if you know how to work with this in other languages, then I think by analogy you can use this version as well. The I/O parameters should be similar
https://radfiz.org.ua/files/temp/Lab1_16/alglib-3.4.0.csharp/csharp/manual.csharp.html#example_mcpd_simple1
here are some examples, search for "markov"
https://radfiz.org.ua/files/temp/Lab1_16/alglib-3.4.0.csharp/csharp/manual.csharp.html#example_mcpd_simple1
found examples here, search for "markov"
The main thing is to make it useful for the models.
Great!
The main thing is to make it useful for the models.
I'll have a look at it. I've been looking at the regime switching + ML for a long time, but haven't done anything concrete yet
I'll have a look at it. I've been looking at the regime switching + ML for a long time, but haven't done anything concrete yet.
In general, what does this Markov chain do? It was described in Perervenko's article as a smoothing function and corrects the result of NS work.
Do you also want to smooth the series? Judging by the order of the record ("regime switching + ML") - before training?
What does this Markov chain do anyway? In V Perervenko's article it is described as a smoothing function and corrects the result of the NS.
Do you also want to smooth the series? Judging by the order of the record ( mode switching + ML) - before training?
the articles above, if not in 2 words
detecting the phase of the market, as an option. switching between trading modes\models
What's up Max? Have you tried to do what I wrote with my data?
What does this Markov chain do anyway? In V. Perervenko's article it is described as a smoothing function and corrects the result of the NS.
Do you also want to flatten the series? Judging by the order of the record ("switching + ML") - before the training?
Holy crap, you guys... what the hell is smoothing? Who are you listening to... You're talking nonsense, ZZ in the targets, now it turns out hmm - smooths out the series))) Markovchain rather mediocre predicts sequences as probabilities of change states, you can certainly adapt, it was even on the site Fischmann Foundation (quantum brains capital), they wrote that they use Markovchain, about 6 years ago. Now they use LSTM.
I read the articles. One also uses it for correction, the other for trading.
Something about Markov chains reminded me of clustering (also without a teacher and you can get several modes/clusters to choose from).
What about Max? Have you tried to do what I wrote with my data?
Not yet, I got caught up in the theory
Because it works a little differently in RL.
Read the articles. One also uses it for correction, the other for trading.
Something about Markov chains reminded me of clustering (also without a teacher and you can get multiple modes/clusters to choose from).
Yes, then detect them in the future and choose a model that works well on this particular cluster