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
I'll hope it comes someday.
All right, good luck. Don't believe the starter!
(Gone to the 'homeland' almost - in 'predictions and consequences...')
Regarding correlation I can say that the analysis requires a criterion for identifying nonlinear correlations of one order or another, and a method for transforming the series so that only linear correlations remain. This can be informally judged by the type of ACF or CF, but I don't recall anything like that on the forum. People don't go any further than Spearman.
Information/copula-based approaches are of course useful, but linear methods are also sufficient to estimate model parameters. Another thing is that analytical form of models turns out to be too complicated, so we have to use iterative methods.
By the way, on this forum it seemed to be a question of information, but that time it all started with finding autocorrelations of conditional second moment.
p.s. Found it: https://www.mql5.com/ru/forum/135430
The forum has been unable to construct a regression line mathematically correctly for ten years, let alone a correlation.
And you don't have to build a regression line! You have to draw a line of progression!
And you don't have to draw a line of regression! You have to draw a line of progression!
And the impossible will become possible!!! ;)
Regarding correlation, I can say that the analysis requires a criterion for identifying nonlinear correlations of one order or another, and a method for transforming the series so that only linear correlations remain. This can be informally judged by the type of ACF or CF, but I don't recall anything like that on the forum. People don't go any further than Spearman.
Don't accuse everyone so indiscriminately. Some people know what ACF is and use it very intensively. I will not vouch for everyone. Personally, I look at ACF in the first place
https://www.mql5.com/ru/code/8295 It must just be hard to find. The code has been in the public domain for 6 years
The forum has been unable to construct a regression line mathematically correctly for ten years, let alone a correlation.
Can you be more specific?
Can you be more precise.
It was described in detail somewhere in my posts on the mcl5 forum, but I was banned, disabled and deleted there for making unflattering statements about the company, so now I can't find it, and I don't want to rewrite the rambling. The point is that the evolution of the residuals FR when moving the line was monotonous, while respecting the minimum norm, without jumps, which can be tracked by identification of FR, however, to build FR on the forum is also not able to. As a simplified approximation - using the minimum of absolute deviations, rather than the square, because in this case the residuals FR evolves more monotonically, but this is not available on the forum either.
Taking into account your contribution to the forum, I can send you an algorithm (code) for constructing a straight regression line by minima of absolute deviations, if you are interested.
ok, good luck. don't believe the topic-starter!
(Gone to the 'homeland' almost - in 'predictions and consequences...')
forDebugger
Speaking for myself, not for everyone (as you like to do).
1. I have never been in any affiliate partnership and never will be...
2. I have called and am calling to run from Forex to the stock exchange. You may think that it is easier to make money there. Forex is the most difficult market.
A misconception that you cannot earn on the Forex market (the theme is called that). You are the only way to earn money in the Forex market.
S.U. Stop trying to present your delusions as the truth in the last instance ... if you can't make money, learn ...