Regression equation
Do you actually want to eat or a recipe (sorry, it slipped out). My opinion, this method describes the selected part of history no better and no worse than others, on forward they behave the same way as the vast majority of systems lose.
More specifically, I am particularly interested in multivariate regression. Looking at options for solving nonlinear regression is also interesting. I have not found any algorithms to deal with multivariate regression in MQL. If you provide me with links and indicators (if you're not too lazy of course) - it will be great! I'm not going to find a holy grail, but to understand the multivariate regression methods from the perspective of time series of currency pairs - it's extremely important for me.
I will be grateful for your feedback.
On point 3, LOC is inefficient for quotes (if you are going to use regression as a predictor). Better to use LAD or quantile regression. It's more complicated (much more coding and science to do), but it works - unlike least squares.
The cause of LOM inefficiency, by the way, is the notorious fat tails. Quantiles do not have this defect.
Can you be more specific?
The MNC is positioned, among other things, as a method for estimating the best parameter choice for a function chosen a priori by the researcher.
Formulas for calculating these parameters minimizing the square of the deviation of actual data from the approximating function have been derived for a set of functions.
Where do the fat tails appear?
Please, enlighten me...
Where do the thick tails arise?
Such a target function -- the sum of the squares of the errors -- is only optimal when the error distribution itself is normal.
Where did I talk about distributions?
Or the topicstarter?
We're talking about an approximation to a polynomial. No more than that.
But no less.
And where is the inefficiency of MNC here?
;)
--
but error studies for normality, an important element of a priori model plausibility estimation...
No argument.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Members of the forum!
Can you advise - has anyone worked with regression equations for currency quotes? Are there any indicators, MTS, MQL-librariesbased on them?
For those who are in the know, could you answer a couple of questions?
I am interested in regression equations. But I face the problem of their adequate description. What data do we have: time (say M15), HIGH, LOW, OPEN, CLOSE, VOLUME. For us it is a set of observations. We have an indicator for which we need to establish a functional relationship with the object parameters (in our case, the change in the exchange rate) - factors. Required: to establish a quantitative relationship between the indicator and the factors. In this case, the task of regression analysis is understood as the task of identifying the functional dependence y* = f(x 1, x 2, ..., x t) that best describes the data we have.
The function f(x 1, x 2, ..., x t) that describes the dependence of the indicator on the parameters is called regression equation (function).
So. Question 1: Of the data we have, which one should we choose as the Indicator and which one should we choose as the Factor? Logically the Indicator is time, the factors are H, L, O, C, V
In our case it is a time series.
The next task is to choose the functional dependency. An equation that describes the relationship between the variation of the indicator and the variation of the factors. Often these are polynomial functions. A particular case is the polynomial of degree 1 - the linear regression equation.
Question 2: What is the best polynomial to choose, and how to adequately describe it in terms of time series, what parameters to apply, what is the degree of the polynomial. Has anyone used Chebyshev polynomial? If so, what is the order?
Our next task is to calculate the coefficients of the regression equation. The usual way is to use ANC.
Question 3: What is the best method for calculating the coefficients for our case?
Question 4. Is it necessary to normalize the data?
And the most interesting question. how to make forecasts of, say, the next tick on the basis of the obtained data and the regression equation?
I would be grateful if someone would share his/her experience, ideas.