a trading strategy based on Elliott Wave Theory - page 277
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
This is an old code (if you compare it with the results I posted), found it in the archive and of course it's not exactly Hurst, that's correct. As I wrote, Hurst for a series of quotes will always (by definition) be around one, it's the nature of it, no matter how you count it.
Just put it out there, without any ulterior motives, maybe someone will find it useful. On the whole it works, if you close your eyes to the fact that it shows values greater than unity.
By hand you get the same "code". No error in it, got what I wanted at first. It was just once the first attempt (recall, after page 30 of this forum) to make it more sensitive to the dynamics of change.
I've been looking at your algorithm for Hearst with my eyes wide open for a long time. Then I got into Peters and so on. Positive moment: finally understood how to count Hirst. Negative moment: I understood the difference of your method of calculation.
This negative moment (as it should be in dialectics) inevitably leads to a positive one. You, Sergey, have proposed a completely new indicator. I propose to call it not Hurst but Grast ! :-))
If, moreover, its use allows to do something in forex, then it is much more valuable than the old Hurst. I mean it. :-|
A lot depends on the period. What period do you trade on?
This negative moment (as it should be in dialectics) inevitably leads to a positive one. You, Sergey, have proposed a completely new indicator. I propose to call it not Hurst but Grast ! :-))
If, moreover, its use allows to do something in forex, then it is much more valuable than the old Hurst. I mean it. :-|
Now in Russian - what do these formulas look like?
Hi Yuri!
During my research, I dug up nine "official" methods of calculating it (some developed by entire institutes) and they are all called Hirst indices, and all gave different values for the same series. Thought a lot.... :о)))
As I wrote earlier, this is where I started my research. In other words, I don't use this algorithm now, but that doesn't mean it's at all bad. If you want it can be normalized, if there is an urgent need that the indicator would give values from 0 to 1.
From memory it seemed to work, and was no worse than many others, at least I checked it with statistics. It looked quite simple: I sequentially built channels, calculated different Hursts, and looked what was in the future. By the way, the classical calculations showed complete chaos in practice, in the sense that they did not correspond to the future in any way. Against their backdrop mine looked quite decent i.e. sometimes no different :o)))).
As warned, sometimes he lies (it happens to many, there is a very long list of things he lies about). It also very important which structure to sample, (I'm emphasizing the important moments :o) then it lies less or doesn't lie at all.
I hope the algorithm will help someone or prompt some thoughts in this direction.
PS: I'm modest, let it still be called Hurst
:o)
to Rosh
Admittedly, I don't really understand what you don't understand about the algorithm. I assure you, the square root I drew by hand will look much worse than in Matcad-e.
to Alex Niroba
Hi Alex! I read your success, congratulations! I strongly recommend, choose the island with Aborigines... It won't be so boring. :о)
[/quote]
Hi grasn .
Thank you. As soon as I've found and bought myself a suitable
island, I'll let you know and we'll have a beer. :-)))
There's no way it's going to work. There is no way a true Hurst can go beyond the interval (0.1). Not because I want to, but by nature. So agree to my suggestion at once. :-))
What do you really care what it's called? As long as it works!
Quite right, the Hearst figure should not go beyond 0:1, but on the other hand a proven (that it is the Hearst figure) and well described wavelet transform calculation (as an example) can literally go beyond that, especially on small samples. I've conceded that it's not really Hurst, I'll have to agree, at least I'll have someone to ask. :о)))
PS: When using it you just have to consider the highlighted word combinations.
to Alex Niroba
ok. I know a couple of great places, they serve great nectar. :о)