a trading strategy based on Elliott Wave Theory - page 72

 
I had this idea this morning. Maybe I did not understand correctly what Bronstein and Semendyaev meant, but I understood that search of an extremum of the energy functional is search of such trajectory function of movement that during the movement the potential energy of the moving point will change as little as possible, ie the faster (without returns) the price has gone a certain way the less its potential energy changed. But if we have chosen the approximation of the equation of the trajectory, what should we do, if we have only one channel of regression for two bars of the beginning and the end and what minimization we may speak about?
this sample may not be built in the only way at the given moment of time
, but it considers a group of channels selected according to the criterion RMS2/3>RMSCO as the same constructed for the given moment of time and among them the minimization should be performed, and in my opinion the channel slope angle is the steeper it is, the lower the price went up and down within it and thus the potential energy changed the most times. All these conclusions may be based on the wrong understanding of the potential energy functional that is why I would like to hear other opinions about it.
 
It's not even the angle, but the path length within the channel, although I may have got it wrong :) as for a group of channels, the path length (relative) may vary insignificantly.... ok, i will wait for what others have to say :)
 
Let's imagine an elongated (totally non-ideal) trough, the bottom of which has a slight convexity and lots of knots and potholes. Putting it slightly at an angle, you get something like a Galton board. Oh! A bobsleigh track might be better. Let's run a bunch of ball bearings down this track, one after the other. They will all roll on their own random trajectory, but stay within the boundaries of the channel. They will all have the same property - the Hamilton function will have close values for each ball. That's the message :)
http://nature.web.ru/db/msg.html?mid=1184545&s=
 
Let's imagine an elongated (totally non-ideal) trough, the bottom of which has a slight convexity and lots of knots and potholes. Putting it slightly at an angle, you get something like a Galton board. Oh! A bobsleigh track might be better. Let's run a bunch of ball bearings down this track, one after the other. They will all roll on their own random trajectory, but stay within the boundaries of the channel. They will all have the same property - the Hamilton function will have close values for each ball. That's the message :)<br / translate="no"> http://nature.web.ru/db/msg.html?mid=1184545&s=


Obozhit, you consider one channel and a bunch of trajectories in it and I was talking about one trajectory and several channels, maybe there is no difference, but it doesn't seem so to me.

ZS Thanks for the link interesting site
 
It seems to me that what Vladislav said can be divided into "technology" and a rationale for his system. For the time being, I assumed that the reasoning about the potential was for the latter. As for "technology", we can assume that the "fan" groups of channels represent one "true" channel. How to isolate it? Solandr has approached this statistically and he may be right.
 
I agree with you that "technology" and reasoning are a bit different things, but without a complete understanding of the methodology we risk a deadlock. Of course, Sounder's results show that everything works even without a field, but why did Vladislav choose this criterion as a basic one?) Let's just say I also built the Expert Advisor "calculation kernel" without the field and now I am busy with trading functions(the absence of the debugger seriously bothers me, yesterday I had a funny story - one and the same Expert Advisor was opening deals on one machine but the same data was not opening on the other one) I've always thought about it but I've never understood it and now I've got a chance to use it, moreover it's not of interest to me only.

PS Actually not absolutely the field theory came to mind... The matter is that in my opinion the quantization of the market is evident (it's pips and lots, support and resistance lines, etc.) and the character of the movement says that it is difficult to predict exactly where the price will be the next second, in my opinion it's very akin to quantum mechanics, but alas my knowledge is almost at an end :)
 
I had this idea this morning. Maybe I didn't quite get what Bronstein and Semendyaev meant...,<br / translate="no">


Who are Bronstein and Semendyaev?
 
Yes, the thought of quantum mechanics begs to be said. About levels - apparently a large proportion of them are just clusters of stops, i.e. subject to the laws of psychology. Since people behave in a high degree of uniformity, an empirical approach would be to obtain data from a representative sample. For example by organizing several banks and DCs in several countries :). And I wonder if this is not what hypothetical "big uncles" are doing? :)
 
Сегодня с утра меня посетила такая идея. Может я не совсем верно понял что имели ввиду Бронштейн и Семендяев...,


And who are Bronstein and Semendyaev?


:) Yes, it's a handbook of higher mathematics, the one I had on hand.
 
<br / translate="no">Candid 06.07.06 18:19
No, the tuner doesn't really load, it's pretty much the same without it. Seems like you can really speed up. Although, perhaps it depends on what is in the terminal (how many and what windows, indicators and EAs). But I'm afraid there is no way to figure it out.


If you haven't seen it, it may be interesting - "MQL4 , JDK1.4.2 and others: comparison of speed".