a trading strategy based on Elliott Wave Theory - page 66
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
Rosh, with all due respect ...
What is this statement ? What is the trajectory of force ? What does it have to do with a quadratic function ?
And most importantly, what can you deduce from this statement ?
Mathematics and physics operate with clearly defined, meaningful concepts. No need to create an inedible mishmash. The simplest analogy is the pendulum. It, too, goes from boundary to boundary. And it is described very clearly both mathematically and physically. So the force there is a linear function, and the quadratic function is potential energy.
A controversial point of view. Besides, the profits are carried by what has already been done before. Potential energy has nothing to do with it. But perhaps its use has the potential :-) to bring big profits. So why not look into it ? Even if only for the love of art ?
By the way, if it's not too much trouble, have a look at the "ZigZag MetaQuotes and other questions" thread.
I asked you something there and am waiting for an answer. And besides, pay attention to Renat's post.
Why not offer MetaQuotes your corrected ZigZag code? There would be a normal indicator in the standard delivery.
...The type of potential field you have chosen is a direct analogue of the gravitational force near the Earth's surface, only with the difference that we take a point instead of a plane (Earth)....
I had a different idea about this... I understood Vladislav did not in vain talked about the events that the market remembers for a certain time, those that over time the force of the influence of events fades, and from here it follows that we have a lot of sources of the field that I think does not attract and push away the price (more precisely, it seems to me the events of the past repel it, and the future events attract it, this indicates (at least I see it ;) When the news arrives, the price moves in a parabola with its branches directed to the place where it was before the news hit the market and when the news arrives, the branches are directed to the expected price position in the future.) It is hard to say how to take it all into account, but I don't think it should be considered so easy.
Rosh, with all due respect ...
What is this statement ? What is the force trajectory ? What does the quadratic function have to do with it ?
And most importantly, what can you deduce from this statement?
OK, let's put it another way: the trajectory of motion will be a quadratic function due to an external force in the potential field that has occurred (there was momentum, which is now neutralised over time).
Mathematics and physics operate with clearly defined, meaningful concepts. No need to create an inedible mishmash. The simplest analogy is the pendulum. It, too, goes from boundary to boundary. And it is described very clearly both mathematically and physically. So the force there is a linear function, and the quadratic function is potential energy.
I have also thought about the pendulum, but have not yet solved the equation to recover the forgotten one.
A controversial point of view. Besides, the profits are carried by what has already been done before. Potential energy has nothing to do with it. But perhaps its use has the potential :-) to bring big profits. So why not look into it ? Even if only for the love of art ?
So suggest your options.
That's pretty much what I've done. The pendulum is the closest model.
Its solution is the equation of harmonic oscillations that more or less corresponds to the price oscillations inside the channel. However, Vladislav correctly suggested that it is wrong to search for the price trajectory, moreover in the analytical form. The process is statistical. Therefore we should not search for the trajectory itself, but for the limits of its fluctuations. For this, it is sufficient to have potential energy in the form of a quadratic function and the trajectory equation in the form of a linear regression. This is what I understood.
And what I did not understand, I expounded in that post on page 26. In short, it sounds like this.
1. Any scalar price and time function will have the property of potentiality if the field of acting forces is represented by its gradient. But we know nothing about the field of forces acting on price.
2. What can we get from potentiality ? how can we use it to select samples if the work of forces along ANY trajectory is the same ?
3. What is the quality criterion of the potential energy functional that can be used to find an optimal sample ?
This is what I would like to think about. After all, it is difficult to trust the result obtained by groping.
It would be desirable to understand what we do and why we do it this way and not that way. What Vladislav can't take away from me is exactly that.
A human searches for channels subjectively, while the machine searches objectively according to mathematically valid criteria. Machine does not care what to count - video playback or search for channels, but for man - it is a lot of stress - watch and search for channels (this is exactly what all traders who do not have this methodology do every day). It takes me about 48 seconds to scan 6400 bars in a sample of M30 period and select required channels on my P4 2.4 GHz machine. Can you identify the channels visually faster than the machine? Moreover, in my Expert Advisor, the results of which I have presented above, I calculate a sample of only 300 bars, which takes less than 1 second including the necessary graphic information on a chart. Such a short sample is taken only to be able to wait for the results of running the Expert Advisor on history. Ideally, I plan to run the Expert Advisor a number of times to find the optimal sample length for searching channels. But this will of course take some time.
...
There is also the following question. Usually we have several channels of different calibre, selected by criteria. A classic variant is 3-4 channels. One is the largest and the others are smaller, which are actually details of the main channel. We can find minimum potential energy points in the above described manner for each channel. Now knowing minimum points of potential energy for each channel how can we use this information to trade? ....
Maybe you will find some other application of potential energy minimum, but Vladislav clearly stated
Vladislav 14.06.06 21:06
Exactly right - I wrote about it, that a minimum of potential energy function serves as one of criteria for channel selection.
That is, he considers this minimum for all channels, not for selected ones, and this criterion with its weighting coefficient participates in channel selection. So to say, the selection that we are doing now does not take this criterion into account and maybe we are selecting the wrong channels or the wrong number of them, although it looks nice and does not have high energies ;)
I have a suggestion on speeding up the search for channels (of course with appearance of error) it is possible not to run through all channels in a row but with some step for example selections differ not by 1 bar but by 5 and in theory it saves time by 5 times and should not greatly influence on calculation accuracy (Duron800 has to use all kinds of tricks ;))
Sorry Solandr read your post of 01.07.06 22:59
...
To what extent is a certain channel just visible visually? Or, in other words, who will find the channel first - the person at the terminal or this algorithm?
...
Please note that the algorithm finds different trend channels on the same history for different final bars.
Let me try to reword your question: "How small is the number of bars needed for this algorithm to detect a channel? How short channels does it detect?".
On a small number of bars, small channels often turn out to be short-lived. For example, a short channel first detected on the previous bar (i.e., ending on it) may be rejected on the next bar due to failure to meet one of the criteria.
The algorithm with the softened minimal size restriction often rejects such short channels by other criteria, but not always.
Long channels are less likely to be shifted, but often change smoothly. For example, the start shifts a few bars to one side or the other .
That is, if you want to decrease the calculation time at the price of quality, you can search not too many bars of history each time. And the deep search can be performed less frequently, holding the results until the next deep search. You can not do away with it at all - channels that are getting out of date should be updated. It is also very desirable to refresh them right after a fast and swing market.
My implementation of search on the same P2.4 machine takes about two seconds per one run for 6400 bars of the previous history together with markup on the chart. On mql there is only the markup.
If slower it would take too long ;-) for the picture which is on page 25, although it is not correct ;-)
Man looks for channels subjectively, and the machine objectively according to mathematically valid criteria.
...
The short sample was taken solely for reasons of being able to wait for the results of the expert's run on history.
Probably in real time such numbers are acceptable (although there is no limit to perfection :), but with history tests, as far as I understand, the situation is quite difficult, even on a short sample. I hope you don't suspect me to be an opponent of the use of computer calculations? :). Just after reading the thread I got an impression that it would be desirable to speed up the algorithm considerably.
Eh:
Only take into account that the algorithm finds different trend channels on one and the same history for different final bars.
Hmm, this, in theory, is not good. However, probably it should be, if calculations start from final bar.
Eh:
...
On a small number of bars, small channels often turn out to be short-lived. For example, a short channel first detected on the previous bar (i.e., ending on it) may be rejected on the next bar due to failure to meet one of the criteria.
...
Long channels change less frequently, but often smoothly. For example, the beginning moves several bars to one or the other side .
...
An update is also very desirable immediately after a fast and swinging market.
If I refer to the algorithm developed here, this experience will be very useful, thank you. But right now I feel like doing the opposite :). The first stage is a short sliding channel. I have already made such an indicator; I even have a picture of it
. The thin lines show interval of 2.5 RMS. It may seem that as long as the price remains within the minimum channel, the previous trend will be in effect and nothing can be done. Only when the price rebounds, we can start looking for a real channel relying on the exit point. However, after messing around with it I might change my mind :).