Bayesian regression - Has anyone made an EA using this algorithm? - page 33
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
Regression analysis does not require a normal distribution of the input data, it requires a normal distribution of the model residuals.
All economic data, price characteristics etc. are correlated. There is no uncorrelated data.
Price is time dependent.
Your post makes zero sense: "Soviet ships sail the expanse of the Grand Theatre; the forex market is a decentralised market; everything is interconnected; everything depends on each other and on time too".
You know, I might as well argue that the Great Porcelain Kettle is circling between Mars and Earth and controls all the markets on Earth...
I don't want to get involved in the bullshit you impose, but rather you should show us your real knowledge on the subject: what real deterministic relationships work in the markets (just don't tell us anything about the correlation matrix, because it's trivial).
The time dependence as a mandatory criterion for the applicability of regression analysis is a bit hilarious, please......
Don't pretend to be an idiot and don't twist my words. Where did I write that time dependence is a mandatory applicability criterion. It's just that this particular thread is attempting a forecast based on a regression model. Or are you saying that for a "tomorrow will be more expensive than today" forecast, the time dependence of the predicted process is not required?
So you want to be shown the correlation coefficient between, for example, price (t-1) and price t and you really don't know that there is a strong correlation between these variables?
Or show the correlation coefficient between e.g. EURUSD and AUDUSD?
Don't you know that there is a strong correlation?
About time dependence as a mandatory criterion for the applicability of regression analysis - tinny at all, glad......
Dimitri:
The price depends on the time...
Yeah, well, we're waiting for your proof of this apogee of your creative thought.
Receive, xxxxxx, a grenade:
EUR Multiple R = .70504504 F = 1654.618
R?= .49708851 df = 1.1674
No. of cases: 1676 adjusted R?= .49678809 p = 0.000000
Standard error of estimate: .076419726
Intercept: 5.857784198 Std.Error: .1120961 t( 1674) = 52.257 p = 0.0000
To decipher or not to decipher?
The independent variable is time.
The dependent variable is EURUSD, D1.
R^2 = 0.49708851
R = 0.70504504
I am amazed by the high level of proficiency in mathematical methods of the participants in the discussion against a complete lack of understanding of the principles of their applicability. Any regression analyses correlated data. If there is no correlation, then the regression is not applicable. If the distribution of the quantities under study is different from normal, parametric statistics methods are also not applicable. The market does not have the property of normality. Also the market as a process doesn't depend on time. Both of these things, however, defeat the very idea of regression analysis at its root.
Vasily, I'm sorry. But this Normality of Distribution stuff is getting on my nerves. Sorry again, for an immodest question, are you zombified somewhere, you are like a copy of the normality of the distribution? Here's only one who manages to codify as opposed to all demagogues.
And I am surprised at the inconsistency in the posts of participants with a high level of competence. Recently, in another Illita thread, you confirmed the existence of a normal distribution. True, there it was about the spread and you wrote: "Distribution analysis is of interest only from the point of view of studying trading conditions. The fish isn't here." And now you write "The market doesn't have the property of normality".
I'm not the one who wrote about the fact that volatility, increments, have a distribution law close to normal and gave charts. I just took it into consideration as I believe in it.
In general, I am interested in the Bayesian approach itself and the attempt to calculate a probability measure as a product of probabilities using Bayes formula. And it's up to everyone to build a regression on it. I think there's a fish to fry here.
Firstly, that a normal distribution exists I have never denied.
Secondly, that post was about data of a different kind - measurements between supply and demand prices.
Thirdly, the process was considered as a first approximation. There was never even any talk of building a robust model for predicting anything. Yes, it is possible to determine the average spread and its distribution around this average using Gaussians at a glance. I.e. to make such an extended SymbolInfo. It should be noted that judging by the level of his competence, the author of the article must be entitled: "How to make an extended SymbolInfo".