Matrix analysis is at rest here. Everything is too non-linear and too close to 100% randomness. If anything can be done, you have to think about fuzzy logic, entropy, neural networks and the like. And chaos theory, but not of Bill Williams' "trading chaos" but of real, "mathematical" chaos, for which much has been worked out - by the way by Russian/Soviet mathematicians.
The market is a positive feedback system. A change in the equilibrium of the system leads to forces which increase disequilibrium - up to a certain point of course - then back again. Like in generators. Only there it's one inertia one force, and here...
this theory explains well the support and resistance lines and the fact that the most stable trends occur in very calm markets.
The market is a positive feedback system. A change in the equilibrium of the system leads to forces which increase disequilibrium - up to a certain point of course - then back again. Like in generators. Only there it's one inertia one force, and here...
this theory explains well the support and resistance lines and the fact that the most stable trends occur in very calm markets.
Cronex писал (а):
Hi all,
Maybe I'm wrong, but I haven't found a thread about the fundamental application of V.Mat. and its sections such as the forecasting of vector of probable chart development, chart continuity at a point or chart convexity theory for forecasting the probable development of market direction.
Is it not discussed in principle, or is it just that nobody has done it?
I think that if you want to perform technical analysis, you cannot do without it. Many questions can be solved graphically. When I was in college, I wrote systems for modeling physical processes on limited empirical data. Quite accurately it turned out and there is an assumption that in this application will also be useful.
Here is an idea: calculate a plausible vector of graph development and restore it to the target by quadratic interpolation and after that calculate divergence from real behaviour to assess direction and quality of the trend.
Hi all,
Maybe I'm wrong, but I haven't found a thread about the fundamental application of V.Mat. and its sections such as the forecasting of vector of probable chart development, chart continuity at a point or chart convexity theory for forecasting the probable development of market direction.
Is it not discussed in principle, or is it just that nobody has done it?
I think that if you want to perform technical analysis, you cannot do without it. Many questions can be solved graphically. When I was in college, I wrote systems for modeling physical processes on limited empirical data. Quite accurately it turned out and there is an assumption that in this application will also be useful.
Here is an idea: calculate a plausible vector of graph development and restore it to the target by quadratic interpolation and after that calculate divergence from real behaviour to assess direction and quality of the trend.
That's what I think too.
I also think the complexity of programming is nothing compared to the complexity of the problem.
To begin with, look in a calculus textbook and ask what the difference between interpolation and extrapolation is.
Itso:
Matrix analysis is at rest here. Everything is too non-linear and too close to 100% randomness. If anything can be done, you have to think about fuzzy logic, entropy, neural networks and the like. And chaos theory, but not of Bill Williams' "trading chaos" but of real, "mathematical" chaos, for which much has been worked out - by the way by Russian/Soviet mathematicians.
The market is a positive feedback system. A change in the equilibrium of the system leads to forces which increase disequilibrium - up to a certain point of course - then back again. Like in generators. Only there it's one inertia one force, but here...
this theory explains well the support and resistance lines and the fact that the most stable trends occur in very calm markets.
Those who are not booming in applied mathematics take a break. The rest prefer to make money out of it. And what is non-linear can easily be reduced to a linear form and back. As an example, you could look at the least squares method - restoring the maximum correlated function over a number of points.
Matrix analysis is at rest here. Everything is too non-linear and too close to 100% randomness. If anything can be done, you have to think about fuzzy logic, entropy, neural networks and the like. And chaos theory, but not of Bill Williams' "trading chaos" but of real, "mathematical" chaos, for which much has been worked out - by the way by Russian/Soviet mathematicians.
The market is a positive feedback system. A change in the equilibrium of the system leads to forces which increase disequilibrium - up to a certain point of course - then back again. Like in generators. Only there it's one inertia one force, but here...
this theory explains well the support and resistance lines and the fact that the most stable trends occur in very calm markets.
Reshetov писал (а):
Itso:
Math. analysis rests here. Everything is too non-linear and too close to 100% randomness. If anything, you have to think about fuzzy logic, entropy, neural networks and the like. And chaos theory, but not of Bill Williams' "trading chaos" but of real, "mathematical" chaos, for which much has been worked out - by the way by Russian/Soviet mathematicians.
The market is a positive feedback system. A change in the equilibrium of the system leads to forces which increase disequilibrium - up to a certain point of course - then back again. Like in generators. Only there it's one inertia one force, and here...
this theory explains well the support and resistance lines and the fact that the most stable trends occur in very calm markets.
Those who are not booming in applied mathematics take a break. The rest prefer to make money out of it. And what is non-linear can easily be reduced to a linear form and back. As an example, you could look at the least squares method - restoring the maximum correlated function over a number of points.Math. analysis rests here. Everything is too non-linear and too close to 100% randomness. If anything, you have to think about fuzzy logic, entropy, neural networks and the like. And chaos theory, but not of Bill Williams' "trading chaos" but of real, "mathematical" chaos, for which much has been worked out - by the way by Russian/Soviet mathematicians.
The market is a positive feedback system. A change in the equilibrium of the system leads to forces which increase disequilibrium - up to a certain point of course - then back again. Like in generators. Only there it's one inertia one force, and here...
this theory explains well the support and resistance lines and the fact that the most stable trends occur in very calm markets.
Now that was genius. I'll shut up....
The method of least squares is one application of linear regression analysis and it will simply tell you that yes - there is a trend. But the trader determines this by eye quite well. So what? He's out again, because usually the trend is noticed when he runs out.
Of course, ISC is much better than all the pseudo-mathematical methods of trend detection, but it is simply not enough. Just non-linearity leads to the biggest jumps and hence to profits.
IMHO it would be better to say so - there are moments when the movement is unlikely, and there are moments when sometimes small changes in the environment will lead to a movement - and the movement can be both up and down with approximately equal probability.
That's what I'd like to talk about, not "boom boom". ...
I remember in electrical engineering theory at university we were taught about transients: we were given a diagram with a bunch of capacitors, coils and resistors, and we used this diagram to plot the transients. The price movement after impulse spikes is very similar to these diagrams. Question: the parameters of the system are not known, is it possible to draw a continuation with a part of the transient picture? At least without taking into account that the system parameters may change at any time?
2 Integer - transients are negative feedback. The system 'dampens' external shocks. That is why they last for a very short time.
In Forex it happens when the trend is already weak and begins to move sideways - there is a fading oscillation.
If the trend was ascending, this happens when there are no more sellers. Some traders with open positions (of course, long ones) decide to close. This results in a pullback. For one small group of traders it is a buy signal - but they are few and the movement upwards is also small. Then another small buy and downwards, but smaller, etc. You can see on the chart that the trend is fading.
If the market was a pile of capacitors, coils and resistors, that would be the end of it. But it's held by people, and they want movement - some want up, some want down, and some (those who haven't opened yet) don't care - as long as they catch the movement at the beginning. And then it all starts again...
In Forex it happens when the trend is already weak and begins to move sideways - there is a fading oscillation.
If the trend was ascending, this happens when there are no more sellers. Some traders with open positions (of course, long ones) decide to close. This results in a pullback. For one small group of traders it is a buy signal - but they are few and the movement upwards is also small. Then another small buy and downwards, but smaller, etc. You can see on the chart that the trend is fading.
If the market was a pile of capacitors, coils and resistors, that would be the end of it. But it's held by people, and they want movement - some want up, some want down, and some (those who haven't opened yet) don't care - as long as they catch the movement at the beginning. And then it all starts again...
Integer писал (а):
What's the problem? Is it the maths or the programming? If in programming, we can combine efforts :-)
Thanks for the offer, but I have no problem with programming - I've been doing it for 20 years :-)What's the problem? Is it the maths or the programming? If in programming, we can combine efforts :-)
I have a problem with mathematics too, but the subject is new - it is difficult to determine the point of application, while the web is full of nonsense and speculations about the historical behavior of the market in the past.
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
Maybe I'm wrong, but I haven't found a thread about the fundamental application of V.Mat. and its sections such as the forecasting of vector of probable chart development, chart continuity at a point or chart convexity theory for forecasting the probable development of market direction.
Is it not discussed in principle, or is it just that no one has done it?
I think that if you want to perform technical analysis, you cannot do without it. Many issues are solved graphically. When I was studying in the institute, I wrote systems for modeling physical processes on limited empirical data. Quite accurately it turned out and there is an assumption that in this application will also be useful.
Here is an idea: calculate a probable vector of graph development, restore it to the target by quadratic interpolation and then calculate divergence from the real behaviour to assess the direction and quality of the trend.