a trading strategy based on Elliott Wave Theory - page 289
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
I am interested in your opinion, colleagues, about the potential of the method to identify hidden patterns and relationships in the studied BP. When comparing these two pictures, somehow one cannot tell at once that "here it is possible to conduct arbitrage transactions on this process and on that one the case of an efficient (martingale) market is implemented...".
Here it would be interesting to compare the wavelet spectra of both series. For real BP they have well pronounced peaks, which do not average to the end even on large time intervals (~1 year). Moreover, the PL spectra of large parts of a given BP (say, its right and left halves) are quite similar. It seems that for a random series this should not be the case, and in the limit with a large number of terms of a random series its PL spectrum should average to a smooth curve without peculiarities. However, I have not tried it myself - it is only a guess.
There is only one way to get the continuation of the curve right - continue it as it will go. :-) All other ways are wrong. Constant continuation may be the least of the evils, but how much it solves the problem we will only know by research.
That's not grouchy. I just don't like this formulation of the problem: continue curve - build wavelet decomposition - use it to build extrapolation of the curve. If the technique of using GP for extrapolation is correct, we should get our own continuation of the curve. If it is incorrect, we should get a correct forecast only if the error in curve continuation compensates for the error in the extrapolation method. Hm ...
:-))) A masterpiece. Few people know where it comes from.
I wonder how you unfold it in time ? After all, at any given moment the coefficients of the GP are not a matrix but a vector. If you make the frames of this movie into CWT matrix pictures, it will turn out quite differently than in a regular movie, where all the frame points refer to the same moment in time.
:-))) A masterpiece. Few people know where it came from.
That's for sure! It may be old but it's a masterpiece. I love masterpieces.
I take a long enough piece of history. I pick a starting point. From it, I take a fixed BP segment (usually 2048 counts). I do a CWT over it. This matrix is the first frame. Then I shift the start point by a small amount (1-10) to make it smooth. I repeat it all again: this is the next frame. And so on for several thousand times or as long as there is enough history.
It's like scrolling through a chart of a piece of price series, adding a new point to it 25 times per second. (The same thing MT does, only much slower). I, on the other hand, am scrolling through the CWT of the current chunk of the series. Just gliding along the story looking at it through a fixed window.
So that there is no hassle, everything is immediately written to an .avi file via a suitable codec, no intermediate data is saved. Without compression the size of the video file will be ~10Gb, and so just a few hundred meters, which is already acceptable. It all takes a long time (1-5 hours) to compute, so it hasn't made a DVD yet.
It's like scrolling through a chart of a piece of price series, adding a new point to it 25 times per second. (The same thing MT does, only much slower). I, on the other hand, am scrolling through the CWT of the current chunk of the series. Just sliding along the history looking at it through a fixed window.
Yeah, cool. It's like looking out the window of a train at the landscape passing by. But not at breakneck speed, but when you can quite perceive the shapes and even how they change in the wind, for example.
It would be interesting to watch.
It would be interesting to watch.
Yes, I don't mind at all giving it a look. I just don't know how to transfer a file of a few hundred MB in size. Well at least ~50Mb after additional compression.
Well, you could post it on some free service of that kind, for example.
We did it here once when grasn posted MatCad.
But in general, don't get your head around it. Of course it would be interesting, but there's no need for that. Somehow it seems to me that an effective way to use the GP should be sought not in the above pictures, but in understanding the nature of the GP and the processes it depicts.
You can use DemoCharge2005 v1.1.1.9 compiling GIF file or SnagIt v8.2.3.
And the hoster, for example, http://rapidshare.com