What is the optimum depth of history for identifying a useful signal?

New comment
 

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 hear it.

 
ZaPutina:

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 hear it.

I think we should distinguish the "optimal depth of history for analysis" (1) from the "optimal history for trading" (2). For (1) - the deeper, the better, while (2) should be found through optimization by the TS itself out of the available (1). Roughly.
 
Yusuf, I'm looking in a different sense, I'm talking about some idea of how to cram a Fourier into it, so that it "works" 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.
 
ZaPutina:

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 hear it.

In the morning.
 
paukas:
In the morning.
Take your hat off (c). Don't insult history.
 
ZaPutina:
Take your hat off (c). Don't insult history.
You asked, I answered. The result is obvious.
And Fourier won't give you anything, just a waste of time.


 
paukas:
You asked, I answered. The result is obvious.
And Fourier will give you nothing, only a waste of time.



You didn't answer, you- you blurted out something, you tried to be witty. You can of course smear that the analysis is done with the opening of the day, week, month, but nevertheless the beginning - in the morning, so in the morning.

But, the same way you can answer the question of how to fix this or that bug, the answer-hands, and right like a witty, though cheap from this wit smells. So hat's off, don't insult history by giving away such cheap feathers under a historical guise.

ZS. To Fourier. Maybe it won't do you any good, I don't know how you've used it. For me this question is far from being so unambiguous.

 
ZaPutina:

You didn't answer, you- you blurted out something, you tried to be witty. You can of course smear that the analysis is done with the opening of the day, week, month, but nevertheless the beginning - in the morning, so in the morning.

But, the same way you can answer the question of how to fix this or that bug, the answer-hands, and right like a witty, though cheap from this wit smells. So hat's off to you, don't insult history by giving away such cheap feathers under a historical guise.

ZS. To Fourier. Maybe it won't do you any good, I don't know how you've used it. For me the question is far from clear-cut.

Apparently he is coming from his own experience. Probably he does so and it should be kept in mind that he trades successfully.
 
khorosh:
He seems to be speaking from experience. That is probably what he does and it should be borne in mind that he is a successful trader.
You are right, everyone comes from experience and has what they deserve.
 
khorosh:
Apparently he's basing it on his experience. Perhaps, he does so and we should keep in mind that he trades successfully.

Don't support this clownery. There was nothing in the answer that had anything to do with the question, just speculation about the starting point, whether it was the morning of the day or the morning of the month... Don't you feel it answers the question about the depth of history for analysis? Maybe, he wanted to say that if you trade on the watch, the period of analysis is a day (approximately), if on H4, it is a week (approximately), on D1, it is a month (approximately), etc. All this is rough, but nevertheless it can be justified. If you do not know what I am talking about, you can ask: What I am trying to do is to open a door, what I am trying to do is to open a door...

ZS. The notion of successful trading has variations too... I don't want to argue about it.

 
ZaPutina:

.... morning of the month...

Now that's really clowning. ))
123456789101112131415161718192021
New comment