Tics: amplitude and delay distributions - page 7
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
Literally two pennies. For practice, not for deriving analytical solutions, it is enough to have two histograms of distributions, the initial one and the resulting one. All you have to do is to divide the range of the initial distribution into irregular "bars" so that you get the one you need. This is easily accomplished with a set if. Accuracy, of course, is not the highest, but in practice, it is often enough to check some ideas. We can go a bit further, find stupid approximation of original and resulting series values by high degree polynomial. This is the simplest method, based on histograms, and it is not universal and subject to "edge" effects. But, it often takes much longer to find an analytical solution and the result is not guaranteed.
Well, something similar to the Kolmogorov test, only not cumulative and used differently. Sorry for the repetition, the goal is very simple, from one histogram (a set of "bars" with given boundaries on the scale of values and the height of the "bars" characterizing the probability of values appearing from the scale of values) get another, this is the root. Without distracting from theory, forms, description, etc. - simply axing the histograms to each other. The method is rather crude.
The Kolmogorov test seems to imply knowledge of theoretical FR. North Wind please elaborate, I'm a little confused. We have two histograms, if PD (distribution laws) are unknown then we are looking for a theoretical curve by some criterion of agreement, say chi-square. If we have found it, we will work with them. If we failed, we should approximate them with a polynomial (we can approximate them not with a polynomial but with polygaus or polyrice approximations). It is necessary to have at least some analytical description of the resulting series (even by polynomials) to get the resulting series from the initial one. If you have analytical description, it is easier to use rnd(1). How to get from one histogram (random variable) to another without approximation? I don't understand :-( Or you're talking about Kolmogorov-Smirnov two-sample test, but it's intended only for test of hypothesis about ZR coincidence and doesn't transform one histogram to another. Mathemat in matcad is the same symbolic computation, the Maple kernel is used
What is rnd(1),-a normal unit distribution?
What is rnd(1),-a normal unit distribution?
Ahh, I see. I guess what Prival wrote : "If you have received an analytical description, it is easier to use rnd(1)" - definitely makes sense. I did not have rnd(1) in my experiments but it should work out at first sight. I'd better let Prival tell you, I'm also interested in principle.
Aha finally found a branch where they are waiting for my grain of knowledge (can barely find it).
I attach the file, there is how to get another one from one table + algorithms how to generate different laws of distribution having rnd(1).
By the way there on pg. 41 see figure 1.10 as constant probability elipse. It seems to me that this is how we should draw the quote chart, not as OHLC (here I told about it 'We need indicator that reflects price in operational time'). I don't know, but it seems so right to me, because the measurement of the value without the accompaniment of accuracy makes no sense.
P.S. Mathemat I will scan all Tikhonov for you :-)
You can work with ticks by setting tick control blocks such as a tick pass in all currency pairs, no matter how many ticks have passed, you need at least one tick simultaneously in all currency pairs, the beginning of this pass is the same as the end, at least one tick has passed, this is the control point, so to speak, a multi-currency tick. One tick, the beginning of the block, the last tick finishes it. The initial tick may be a single tick for one currency pair; the last tick must be a single tick for another currency pair; during this time, from the initial tick to the final tick, many ticks in other currency pairs can pass. It's exactly in this multicurrency tick that we should look for some side facts. Besides, we can take any other symbol and attach it to the multi-currency tick, for example, by analyzing in the range of the multi-currency tick or by extending the range of the multi-currency tick.
More details))
(Can you be more specific?))
Did you look at the date of the post?
Did you look at the date of the post?
has the MT4 servers changed the algorithm for generating ticks around the smoothed average external feed?
;)
has the MT4 servers changed their tick generation algorithm around the smoothed average external feed?
;)
And did you try to take not one but 200 or more ticks for the generation algorithm?