From theory to practice - page 523
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
You build a graph, then select the graph and right-click and select "Add Trend Line" from the context menu.
In the window that opens select "Polynomial".
I will highlight the question separately:
What function to do the regression on, so that its last point is in the centre of the price channel?
We need a function, like a polynomial, that can handle both the zigzag and the half circle (as you can see from the examples, the polynomial can't handle these topics).
(Maxim Dmitrievsky's figure can not be considered yet, it does not fit the theory that the price is a trading channel that follows a certain trajectory. this figure is a trading channel with an ejection at the end, it can be considered later).
Other examples of non-linear functions are exponential functions, logarithmic functions, trigonometric functions, power functions, Gaussian functions and Lorenz curves.
Smokchi Struck:
@Олег avtomat вы показали один случай из многих.
даже машка в каком-то случае может хорошо отработать.
покажите хотя-бы 10 случаев разворотов тренда.
Oleg avtomat:
Right. And it should not be on history, but in relation to the current situation. Live developments, so to speak.
Stay tuned. (Well, I will not endlessly clutter up someone else's thread with my pictures).
Forex will provide many cases, -- both ordinary and unusual.
There is a contest (for the whole September), in which the demonstration of live events is compulsory. Description of the trades. Investment password.
Ejeli interesting, I will give the link in the PM.
I will not duplicate it innews, it is too boring, as it turned out...
I will highlight the question separately:
What function to do the regression on so that its last point is in the centre of the price channel?
Unfortunately, there is no such approximation in Excel.
To solve this question, the approximating function should be taken on the condition that it passes through the last point. The function thus changes significantly and it is necessary to apply ANC to this function. Some time ago I solved this question, even made an indicator... I'll look for it, maybe I'll find it...
UPD: I found it.
For example
This indicator identifies price extremums over a certain period of time and then builds an MOOC channel so that it passes through the last points found.
The solid line represents a parabolic (of the second order) and the dotted line represents a cube (of the third order). Both of them pass through the last selected point.
I will highlight the question separately:
What function should I use to do the regression so that the last point is in the centre of the price channel?
I will highlight the question separately:
What function should I use to do the regression to make its last point be in the centre of the price channel?
I need a function, like a polynomial, that can handle both the zigzag and the half circle (as you can see from the examples, the polynomial cannot handle these topics).
(Maxim Dmitrievsky's figure may not be considered yet, it does not fit the theory that the price is a trading channel that follows a certain trajectory. this figure is a trading channel with an ejection at the end, it may be considered later).
Other examples of non-linear functions are exponential functions, logarithmic functions, trigonometric functions, power functions, Gaussian functions and Lorenz curves.
You've already been told it's a waste of time.
Or, like Volchansky, you can build channels via regression and trade on them intuitively. The result will be intuitive too, but nice looking.
Or you may go here https://www.mql5.com/ru/forum/224374
The Hilbert-Huang method is a help, it was discussed here a long time ago and there is an article.
Thank you very much, Bas)) Vistog will have it all.
That's if you spit on the price and use only time and nothing else. (Well the price is only for the direction of the deal and that's it).
I.e. I know that events alternate, now we are in one state and waiting for the transition to another state - the occurrence of a new event.
The potential is there. We just need to develop it well.
forecasting is doomed, because markets are "live" and it is impossible to guess who is waiting for what and who wants to get what
SSA itself is interesting, you can use it to try to assess whether the market has changed compared to the previous one
https://ch.mathworks.com/matlabcentral/fileexchange/58967-singular-spectrum-analysis-beginners-guide
Well, wavelets show the same pictures, but the essence is not to predict - all the same "guessing" will turn out, but to try to find differences in market conditions with the help of several mathematical models, I study SSA, who may be using regression - although there should be a lag, or rather inertia
I don't know, it seems to me that if Alexander failed until the end - nothing will save this approach. Predicting a random walk is hard, and estimating its imaginary states is difficult :)
Oh I don't know I don't know, it seems to me that if Alexander has failed until the end - then nothing will save this approach. Predicting accidental wandering is hard, and evaluating it some imaginary states :)
And why is wandering random? It wanders back and forth. And if it's there now, it's time to come here.