What is the optimum depth of history for identifying a useful signal? - page 3
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
Apparently he is coming from his own experience. Probably he does so and it should be kept in mind that he trades successfully.
What about me?
So what are you clinging to, he has an indulgence for the non-kosherness of Fourier in analysis, given to him and his fangroup. For him, the application of Fourier in analysis is more unkosher than the lard in Paukas.
Feel the scale )))))
What about me?
What about me?
What about your words)))
" It has been correctly written here that the Fourier transform only applies to periodic functions. But still there are those who want to pull it on forex."
So what is the optimum depth of history to analyse. I have my own opinion, but would like to hear it.
Yusuf, I'm looking in a different sense, I'm about some idea of how to cram a Fourier transform in there to make it "work" on a non-stationary price series, with a prior reduction of the process to quasi-stationary, that is as if the Fourier itself is only needed for decomposition.
Fourier is only suitable for periodic series. For aperiodic series it will more or less approximate only the -pi/2 to pi/2 segment, i.e. midpoint, while edges will tend to the value of zero harmonic.
Some wavelet transforms can be used for non-stationary series. The wavelets themselves are selected according to the quality of the restored series after the transformation.
What about your words)))
" It has been correctly written here that the Fourier transform only applies to periodic functions. But still there are those who want to pull it on forex"
One does not contradict the other.
In its pure form the FFT is not suitable for the market. To be honest, I don't know what it's good for at all. There was even a time when I thought it was wrong. Now I think it's just not finished. If you manage to finish it, you might get some very interesting results.
If you are seriously interested in FFT, you should study it not from books but at a higher level. You have to understand the physics of what's going on.
Which frequencies (periods) are in the FFT and which are not and why? Why at even N the spectral counts on the left and on the right of Fd/2 are complex-conjugate? Why at even N the spectral counts 0 and Fd/2 have no imaginary part? What is obtained at odd N and why? Most people, even those who consider themselves experts, do not know the answers to these questions.
If you can answer the above questions correctly, more serious questions should arise that should suggest a way of solving the problems. I am in the process of doing this and so far the prospects look very tempting.
It is clear that the detection of a signal on different TFs will correlate with each other, i.e. signal/noise level will differ on different TFs, therefore it is better to take inter-timeframe signal/noise level, i.e. to catch the dominant harmonic on each separate TF, and then obtain one common for these two (for example) TFs.
Some people find a useful signal (let's disregard demagogy of its existence) on a longer history, some people need 5000 bars and others need much more. For example, if we analyze the depth of 1000 bars on the maximum (which is used in the analysis) TF, then for the minimum TF it will be tens of thousands of bars.
So what is the optimal depth of history for analysis? I have my own opinion, but I would like to listen to it.
What is a useful signal? This question is for all the discussants and sympathizers.
What is a useful signal? A question for all the panelists and sympathisers.
It is quite simple. Signals are useful, harmful and useless.
Earned is useful, not earned is useless, lost is harmful.
One does not contradict the other.
In its pure form, the DFA is not good for the market. To be honest, I don't know what it is good for at all.
It's quite simple. Cues come in useful, harmful and useless.
Earned is useful, not earned is useless, lost is harmful.
Almost right, take a pie off the shelf.
A useful signal is one that can bring benefits, earnings.
So why the fuck look for it on a 5000 bar history?