Machine learning in trading: theory, models, practice and algo-trading - page 2547
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
Shapelet searches are like clustering row segments. Probably useful for signals like cardiograms, but not sure about usefulness for price studies.
By the way, did you manage to figure out how to use the LGBM model? If you trained in R, you could try to use the library of San Sanych).
Shapelet searches are like clustering row segments. Probably useful for signals like cardiograms, but not sure about usefulness for price studies.
By the way, did you manage to figure out how to use the LGBM model? If you trained in R, you could try to use the library of San Sanych).
The problem is specific. I have macbook m1 and wanted to train models without virtualization, but catbust has not been delivered for this architecture yet and lgbm is available. But there you should additionally put cli version to save models in cpp or if/else format. Or parsing python code into µl. It turns out that for now it's more convenient to use winds virtual machine with catbust (they are promising version for m1 for a year).
Is it for testing? Probably I would use win VPS and it would be better not to use pure mql. So, I have to use if/else or wait for promised ONNX in mql)
What's a sanych library?
It's this one. But for some reason it seems that you wrote about it (maybe confused with elibrarius).
This one. But for some reason, it seems that you wrote about it (maybe I'm confused with elibrarius).
Aah, So it's not his, bible he's just a regular user...
It doesn't matter who's the author. The main thing is that it works (with corrections given in comments). The R session is initialized there and you work with it as long as you need, which is useful if you need to keep heavy model in memory (without loading/unloading it at every calculation). In C# the official integration with R is done in a similar way.
This is for testing, right? Trading is probably on win VPS, and it's better not to go beyond pure mql. So, I have to use if/else or wait for the promised ONNX in mql)
Or rewrite the models code in mql, but then save it in c++, to simplify it. It is extra steps.
If VPS is not metaquat, then it is possible to try to compile c++ into dll. In practice, though, did not test this approach.
To be honest, I don't quite understand the question.
Maybe that's what it's about?
I've already read a lot of books.
More specific question, what kernel length h and g from the screen, since you give an example about filters?
Wavelets are the same as Fourier. There is classic Fourier, there is window Fourier, and there are wavelets, where instead of a rectangular window as in window Fourier, windows of a special kind are used - wavelets. For financial quotes, Fourier is not suitable, because of the random nature of the quotient.
Turning to the book, we should probably start with the basics that wavelets are differentiated by the type of transformation.
Continuous wavelet transform.
Discrete wavelet transform.
Discrete transform, divided into two more subsections.
Google what continuity and discreteness are, understand it
and relate it to the world of our subject, to understand which type suits us.
This is probably where we should start to study wavelets.