Machine learning in trading: theory, models, practice and algo-trading - page 2674
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
But it's a fact that you need to filter your awareness.
Unfortunately these are very rarely repeated or hardly repeated spectra, moreover with different behaviour afterwards. All decompositions in physics are used to find previously found spectra in media, in the understanding that the spectrum guarantees some state of the medium. This guarantee is not present here. I repeat, I see decomposition into the simplest functions only in dynamics, to understand the behaviour of these sinusoids, how they decay, which ones are no longer there, which ones have appeared, which ones are a consequence of strong factors, which ones are a consequence of weak factors and, accordingly, noise.
is a closed system with 0 sum, where is there BLJAT heterogeneity there? When one place lost in one place, another place gained in another and nothing disappeared. When there are external (from other markets) inflows/outflows/inflation the denominations are normalised, this is also part of the bargaining.
At the same time physically the whole system and its individual parts are regular like clockwork, because it works according to it, just like you.
Another thing is that it is extremely discrete, there are very few participants, which is generally speaking sudden and unexpected.
I apologise, I wasn't restrained :-)
It is possible to calculate waves in inhomogeneous, knowing the topology of the medium))))))
The nature of the process is of course different, but similar in appearance) Therefore, apparently some methods of physics work, badly, but work).Unfortunately these are very rarely repeated or almost not repeated spectra, moreover with different behaviour afterwards. All decompositions in physics are used to search for previously found spectra in media, in the understanding that the spectrum guarantees some state of the medium. This guarantee is not present here. I repeat, I see decomposition into the simplest functions only in dynamics, to understand the behaviour of these sinusoids, how they decay, which are no longer there, which have appeared, which are the consequence of strong factors, which are the consequence of weak factors and, accordingly, noise.
The whole theory of filtration, both linear (Kalman) and nonlinear generalisation (Stratonovich) is done for a dynamical system. This means a simple thing - knowledge of a mathematical law describing the evolution of the system with time. If such a law were defined for the market, it would immediately mean the creation of the grail (well, or proof of its impossibility)
Any mathematical formulas are always derived from some assumptions, which are not necessarily always and everywhere fulfilled.
Any mathematical formulas are always derived from some assumptions, which are not necessarily always and everywhere fulfilled.
You don't need to know the law of the filtered object to filter....
Well, the very existence of such a law is absolutely necessary. But without knowing its exact form, you can sometimes do without it.
Here, for example, are the assumptions underlying the Kalman filter. The possibility of separation into signal and noise is assumed from the beginning, not proved. For some physical object, these are obvious assumptions, but not for prices.
Of course, nothing prevents us from decomposing a price into components, but no component will be noise (a stationary independent process). Well, in my opinion, such decompositions should not be based on frequency separation, but on the separation of the low-amplitude component. Because of the presence of pulses, these are different things.
closed system with 0 sum, where is there BLJAT inhomogeneity ?
Inhomogeneity in supply and demand (similar to the concept of density). The market in this sense is like a swamp, when some steps stand on a solid surface, while on some other steps the foot falls through. So the problem boils down to determining the parameters of a wave in a pie of materials of different densities, like concrete, sand and water. These parameters are not fixed, but constantly changing, but we calculate these changes naturally with some lag in the size of the window part. Moreover, unlike a swamp, which can fail a person walking only downwards under the force of gravity, in the market there will be an active trampoline periodically under the foot, which will throw the poor person upwards. So what kind of homogeneous environment is this?
At the same time, physically the whole system and its separate parts are regular as a clock, because it works according to them, just as you do.
Yes, but it only concerns activity in general, not direction, which is what everyone is interested in.
Of course, nothing prevents us from decomposing the price into components, but no component will be noise (a stationary independent process). Well, in my opinion, in such decompositions you need.....