Machine learning in trading: theory, models, practice and algo-trading - page 2272

 
Valeriy Yastremskiy:

About the same. How to prepare the data for training.

Well go ahead! If you have any problems contact ...

I will add ))

 
Rorschach:

Ideally, you should make the series stationary, find cycles on the spectrum, build a filter for these cycles.

How to make the series stationary? Correction for the average volatility by the time of day? Logarithm and filtering?

How do you build a spectrum? The deeper the history is, the stronger the noise is, the weaker the signal can be. But everything changes on the market. If we take a short time span, there is no guarantee that it is a cycle and not a random coincidence.

I showed you my statistics on the spectrum.

The entire multibillion-dollar market is full of nocturnals and scalpers, and you say ... some stationary filters, I'm not strong. We don't need a perpetual motion machine, let it change, but not immediately

https://github.com/balzer82/FFT-Python

there is more of this, but I don't understand what he did

https://github.com/snazrul1/PyRevolution/blob/master/Puzzles/DSP_For_Stock_Prices.ipynb

balzer82/FFT-Python
balzer82/FFT-Python
  • balzer82
  • github.com
This tutorial covers step by step, how to perform a Fast Fourier Transform with Python.
 
mytarmailS:

Well, go ahead! If you have any problems, ask ...

I will add ))

No, not like that. If there's a problem, solve it.
 

Regarding Fourier, the way I see it is this. First some decomposition is done, e.g. stl

then loops are searched through bpf

then the loops are wrapped with trade logic. (including MO) It does not look complicated.

Time Series Decomposition & Prediction in Python
Time Series Decomposition & Prediction in Python
  • 2019.07.22
  • s666
  • pythonforfinance.net
[latexpage] In this article I wanted to concentrate on some basic time series analysis, and on efforts to see if there is any simple way we can improve our prediction skills and abilities in order …
 
Maxim Dmitrievsky:

Why are you picking on Fourier?

 
mytarmailS:

Why are you picking on Fourier?

You do it, don't ask too much.

i'll do it soon

Uh... what's with the Fourier when you can use autocorrelation?

 
Maxim Dmitrievsky:

Uh... why do we need a Fourier when we can use autocorrelation?

where here?

 
mytarmailS:

where in here?

how did you look for cycles via PCA or Fourier?

I started doing so much at a time that I got confused... but it's temporary )
 
Maxim Dmitrievsky:

how did you look for cycles through PCA or Fourier?

what cycles? when? in prices?

 
mytarmailS:

what cycles? when? in prices?

where are you at all? in prices, yes ) above it says