Obtaining a stationary BP from a price BP - page 13
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
Yeah, we've got a lot of jokers around here.
Hi Sergey. :о) Good to see you. Where have you been? What interesting things did you study?
Wait, let's take it one step at a time. We have a problem of series transformation, with specified properties:
Suppose you are given such a series and told that it has properties (1), (2), (3). For property (3) you are given a transformation mechanism. By what criteria would you check them?
Can I insert? ;)
Check for outliers ... If there are outliers, then the transformation does not work. If not, check for correlation between the original series and the new series, the information component of the features of the series is preserved. Soros rests ... %)
Can I put it in? ;)
Check for samples ... If there are outliers, then the transformation does not work. If not, we check for correlation between the original series and the new one, the information component of the series features is preserved. Soros rests ... %)
Understand correctly, the possession of the above properties is not a sufficient condition for profitable EAs. For example, Fourier coefficients are stationary, if I am not mistaken - proven, (not a professional mathematician) exactly for this reason this transformation is often used in control systems (not to be confused with TA as some do, there principally other approaches and complete self-sufficiency :o), but as you probably guess, not very and not always helps for TS :o)
Understand correctly, the possession of these properties is not a sufficient condition for profitable EAs. For example, the Fourier coefficients are stationary, if I am not mistaken - it is proved, (I am not a professional mathematician) exactly for this reason this transformation is often used in control systems (not to be confused with TA as some do, they have principally different approaches and complete self-sufficiency), but as you probably guessed it does not really and not always help for TS :o)
Well having a BP which represents the price or its change, extrapolate it. What prevents you from getting an incremental sign on the next bar? Yes you may not get a TS that will bring you profit (I'm not an expert in TS).
Ok if the Fourier coefficients are stationary, what does that give you in the case of non-stationary BP?
Well having a BP which represents the price or its change, extrapolate it. What prevents you from getting an incremental sign on the next bar? Yes you may not get a TS that will make you a profit (I'm no expert in TS).
Ok if the Fourier coefficients are stationary, what does that give you in the case of non-stationary BP?
It's no big deal:
https://forum.mql4.com/ru/24888/page5
You must understand that possession of the specified properties is not a sufficient condition of profitability of Expert Advisors. For example, the Fourier coefficients are stationary, if I'm not mistaken - proven (not a professional mathematician) exactly for this reason this transformation is often used in control systems (not to be confused with TA as some do, they have principally different approaches and complete self-sufficiency :o), but as you probably guess, not very and not always help for TS :o)
Are you dreaming about me at night? ))) You've really gone off the deep end. We're not having an argument. There's nothing to argue about. I didn't come up with the definition of TA. Is it my fault that TA is defined by OBJECT of analysis and not by METHODS?
You can sneer as much as you like, say that my knowledge is insignificant (on what grounds, let me ask?), etc., but what is is what is.
TA - predicting future price changes based on analysis of past price changes.
It wasn't my fault >>!!! He came up with it on his own! I mean, I didn't formulate the definition. It's historically constructed. And what you're doing fits within it.
===Sergey, maybe we should stop this pointless and completely unconstructive argument, eh? And if only argued about something essential - about mathematical models - but no! - We're just arguing about something stupid, about words.
I propose peace. All right? (At the same time we will sleep well.))
nothing special:
https://forum.mql4.com/ru/24888/page5
What is the best way to extrapolate from the average Ak0? I "wrapped" the harmonics on a line drawn through Ak0[0] and Ak0[1]
What is the best way to extrapolate from the average Ak0? I have been "winding" the harmonics on a straight line drawn through Ak0[0] and Ak0[1]
What is Ak0?
Well the DFT generates 2 coefficient arrays for sines and cosines + the average value of Ak0. Since we are using DFT on each sample, in fact Ak0 is an LMA with period = window size. Accordingly, it is necessary to extrapolate the muwings in order to then reconstruct harmonics around them
Are you dreaming about me at night? ))) You've really gone off the deep end. We're not having an argument. There's nothing to argue about. I didn't come up with the definition of TA. Is it my fault that TA is defined by OBJECT of analysis and not by METHODS?
TA - predicting future price changes based on an analysis of past price changes.
It wasn't my fault! It came on its own! I mean, I didn't formulate the definition. That's the way it's historically worked out. And what you are doing fits into it.
===What are you!!!!! Look at your avatar from the outside - God forbid you should dream about it. I just want to set you on the right path. That's the wrong definition. A more correct one is given, for example, on this site in the TA section (which I wrote about). There are simply established definitions that don't need to be changed and tucked away with who knows what. Moreover, none of the instruments (including yours, given on this mite) do not fit the definition given in this fucking wikipendia (there is simply no real analysis of price behaviour and moreover no regularities what it talks about). Anything that has to do with price falls under this definition at all. For example, FA (yes, yes, it does), perfectly self-sufficient stochastic financial mathematics, described for example by Shiryaev in two volumes (facts and models), "stochastic control systems with fixed/random structure" which are also self-sufficient. All of the above work with price series, but work on completely different principles than TA. different this.
You can sneer all you want, say that my knowledge is insignificant (on what grounds, may I ask?), etc., but what is is what is.
no-no-no! not the ego! By the way, I never said anything about your knowledge or called you names ... :о) And how do I know what your subconscious hides?
Sergey, maybe we should stop with this pointless and completely unconstructive argument, huh? If only we were arguing about something substantial - about mathematical models, for example - but no! - About rubbish, about some words.
If it didn't matter at all - would have forgotten long ago. I am in favour of orderly
I propose peace. OK? (At the same time we will sleep well.))
GOES!!!! AGREED!!! NOT ANOTHER WORD!!! I'm generally peaceful, maybe a bit nerdy, but peaceful. :о))))
the world.
Well the DFT generates 2 arrays of coefficients for sines and cosines + the average value of Ak0. Since we use DFT on every sample, in fact Ak0 is an AGR with period = window size. Accordingly, we need to extrapolate the muwings in order to then reconstruct harmonics around them
Uh-huh, senx.
There is only one point, as PF is an approximation. That is, it will give an analytically expressed function for BP in the segment where the calculation will be carried out.
With any extrapolation of this f-function, you will get its form in the future, but without taking into account the changes in the process itself.