Machine learning in trading: theory, models, practice and algo-trading - page 2990
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
And what's not to your liking about digital learning environments))))
DSP has quite a lot of complex maths that many of its fans don't know and which has little application to the market. We all know very well that it always boils down to "let's decompose the price into a Fourier series". This was the case back in the days of Cyberpower, for example).
As a rule, none of these comrades knows that it is not the price that should be decomposed into Fourier series, but the ACF. And those who do know, understand why it makes no sense in the case of prices.
DSP has quite a lot of complex maths that many of its fans don't know and which has little application to the market. We all know very well that it always boils down to "let's decompose the price into a Fourier series". This was the case back in the days of Cyberpower, for example).
As a rule, none of these comrades knows that it is not the price that should be decomposed into Fourier series, but the ACF. And those who do know, understand why it makes no sense in the case of prices.
DSP - oscilloscope, soldering iron, Fourier series.
USSR - Stalin, Beria, gulag.
that's all the understanding...
DSP has quite a lot of complex maths that many of its fans don't know and which has little application to the market. We all know very well that it always boils down to "let's decompose the price into a Fourier series". This was the case back in the days of Cyberpower, for example).
As a rule, none of these comrades knows that it is not the price that should be decomposed into Fourier series, but the ACF. And those who do know, understand why it makes no sense in the case of prices.
Just kidding))))) Of course, outside physics signal processing by static formulas has no chance, or when variables are not properly described or formulated, as in our case just, a person or a group of people has not yet been formalised, signal processing of their actions has no chance yet.))))))
DSP - oscilloscope, soldering iron, Fourier.
USSR - Stalin, Beria, gulag
that's all the understanding...
Well, here it is as if to understand to what precision the variables should be formulated, that the signal from these variables would be significant, with respect to people, how to evaluate it. So far, digitisation of randomly selected signals gives outliers, the same renaissance and other companies. But there's no science yet.
Well, here it is as if to understand to what precision the variables should be formulated, so that the signal from these variables would be significant, with regard to people, how to estimate it. So far, digitisation of randomly selected signals gives outliers, the same renaissance and other companies. But there's no science yet.
To begin with, you need to understand that 99.9% of indicators that are used, including DSP haters, are artisanal/inefficient low-pass or high-pass filters....
They are doing DSP artisanally, but at the same time they are hating it... And all "ingenious ideas" on the use of indicators (artisanal LF, FHF) that can come to the mind of an experienced indicator-user have already been realised in the best way 70 years ago....
If there was at least this understanding, it would be nice...
I'm not talking about understanding that neuronics is the same digital filter, only more complex,
and I'm not talking about the very multi-organ meaning of "filtering".
DSP - oscilloscope, soldering iron, Fourier.
USSR - Stalin, Beria, gulag
that's all understanding...
There's also the great, all-explaining Kotelnikov's theorem. It can also be written down in the USSR achievements.
Instead of grumbling, you can, for example, ponder why Kotelnikov's theorem does not make sense for prices.
to begin with it is necessary to understand that 99.9% of indicators that are used, including DSP haters, are artisanal/inefficient low or high pass filters....
They are doing DSP artisanally, but at the same time they are hating it... And all "ingenious ideas" on the use of indicators (artisanal low-pass, high-pass filters) that can come to the mind of an experienced indicator-user have already been realised in the best way 70 years ago....
If there was at least this understanding, it would be good.....
I am not talking about understanding that neuronics is the same digital filter, but more complex,
and I'm not talking about the very multi-organ meaning of "filtering".
DSP works with signals that are described by stationary (at most quasi-stationary) random processes. Prices are not described by such models. If prices were stationary, all traders would have become trillionaires a long time ago - it's a simple thing to keep in mind.
Filtering is not a prerogative of COC, it uses a special case of this science, not always suitable for prices.
DSP works with signals that are described by stationary.....
DSP is a huge field of digital information processing...
And you wrote about it in your own very narrow sense of understanding.
It's like writing that data science works with signals, which are described by stationary......
To clarify... a signal is NOT a radiogram from an oscilloscope, but any information in digital form.
DSP works with signals that are described by stationary (at most quasi-stationary) random processes. Prices are not described by such models. If prices were stationary, all traders would have become trillionaires long ago - it is a simple thing to keep in mind.
Filtering is not a prerogative of COC, it uses a special case of this science, not always suitable for prices.
And what is the difference between DSP and Mo, well, signal processing by discrete logic with weights and feedback, you can implement a couple of layers on triggers. As if the essence of the argument is not clear. Of course, in traditional variants 80s tsos will not work. But nowadays MO is also used for processing physical processes, and how is it not digital signal processing?
And what is the difference between tsos and mo, well signal processing by discrete logic with weights and feedback, a couple of layers can be implemented on triggers. The point of the argument is not clear. Of course, in traditional variants 80s tsos will not work. But nowadays MO is also used for processing physical processes, and how is it not digital signal processing?
BINGO