Machine learning in trading: theory, models, practice and algo-trading - page 1757

[Valeriy Yastremskiy]

Aleksey Nikolayev:
they say a tinfoil hat protects against spectral entities

If it explained, it would not be worth it at all)))))

[mytarmailS]

Aleksey Panfilov:

I may come in handy:

Thecode base has indicators that calculate the trigonometric phase (in degrees) of the intended wave.MT4,MT5.

The digitized sine wave is easier to move (to select the phase).

Tell me if there is a "standard" sine wave in my ... If you know the algorithm well, it's very easy to move it in degrees, but in my ...

Tell me if there is a possibility to "standardize the price". By standardization I mean

if there is a real time mode, adaptively modulate the amplitudes and change the frequencies so that the time series always has the same characteristics, one spectrum, i.e. expand or contract the amplitudes and frequencies of the graph so that in the end it always has the same characteristics

[Aleksey Panfilov]

mytarmailS:

Alexey, I see that you are well versed in the subject, since you wrote such a complicated indicator...

Tell me if there is a possibility to "standardize the price". By standardization I mean

2. In the real time mode to adaptively modulate the amplitudes and change the frequencies so that the time series always have the same characteristics, the same spectrum, ie to widen or narrow the amplitudes and frequencies of the graph, so that in the end it always has the same characteristics

1. Don't exaggerate. Read the subject:difference calculus, examples. It's simple enough.

2. There are a lot of unfamiliar letters.

And in general you can do everything. It's just long, and there are a lot of ways.

For example, I recently threw in a sentence

The value changes in ascending order. The bar cutoff step can be determined by the phase difference.

"A graph of the equal delta of the trigonometric phase." Might be of interest as the subject of a new article?

.

[Rorschach]

It is interesting to read in DSP (which is sort of about stationary sinusoids) methods of dealing with non-stationarity. It also mentions the logarithm of the series.

It is interesting to compare this with methods from econometrics.

[mytarmailS]

Aleksey Panfilov:

2. a lot of unfamiliar letters.

That is, there is constant normalization by frequency, amplitude and possibly by phase in real time mode.

Do they do this at all? If so, in what tasks and for what purpose?

[Aleksey Panfilov]

mytarmailS:

That is, there is a constant normalization by frequency, amplitude and possibly by phase in real time mode.

Do they do this at all? If so, in what tasks and for what purposes?

)))

I can not assert, but most likely do. A variant of just was proposed in the previous message.

I don't want to get the hang of it, but I'd like to see how a graph drawn on a sine wave.)

In the code base there are indicators that calculate the trigonometric phase (in degrees) of the supposed wave,MT4,MT5. The value varies in ascending order. The bar cutoff step may be defined by the phase difference.

"Equal delta trigonometric phase graph." might be of interest as the subject of a new article?

[Valeriy Yastremskiy]

mytarmailS:

it's not simple, it's primitive, and you're always in a lag, it's like trading a car.

the chain reaction calculation is also primitive, multiplied and added, and a nuclear bomb is made))))Fourier decompositionis also done on primitives, regression analysis, moving averages are just multiplied and added.... well, sometimes logarithms diffuse and integrate, but only sometimes.... Oh, and I forgot the sines....

lagging by one or five minutes is not as bad as by an hour or a month))))

[mytarmailS]

Valeriy Yastremskiy:

calculation of chain reaction is also primitive, multiplied and added, and a nuclear bomb was made))))Fourier decompositionis also done on primitives, regression analysis, moving averages is just multiplying and adding.... well, sometimes logarithms diffuse and integrate, but only sometimes.... Oh, and I forgot the sines....

lagging by one or five minutes is not as bad as an hour or a month))))

well, go ahead ))

[Valeriy Yastremskiy]

mytarmailS:

then go ahead ))

just a question of how to screw high-frequency data to low-frequency phenomena and mona go ahead)))))

[mytarmailS]

Valeriy Yastremskiy:

the only question is how to screw high-frequency data to low-frequency phenomena and mona forward)))))

well, if you formulate the question, then maybe there is someone who can help