Hearst index - page 6

 

hello!!!

Can you people tell me if it's possible to implement this algorithm in C++?

the thing is, I have a term paper on this topic....

 
Is it fair to say that if a series of quotes is characterized by an Hirst value much lower than 0.5, then the tactic of opening positions against the outliers will be effective, assuming a high probability of coming back to the average? And vice versa, if H is considerably higher than 0.5, then a trend tactic should be used?
 

Yes, this is true.

It can be shown that there is an unambiguous relationship between the Hurst index and the autocorrelation coefficient. It's the same here: <0 - rollback tactic, >0 - trend.

 

The Hearst figure is a good thing, but you have to be very careful with it. In its essence, it shows the dynamics of the behaviour of the increments of the analyzed series. In one case, the "general vector" of increments is unidirectional and the series is likely to move away from its current average; in another case, on the contrary, the increments are such that the series will tend towards its average; in the third case, the increments are absolutely random and the series is not predictable. It does not say anything about the direction of the series, the probability with which it will go somewhere, how long it will "go somewhere" and where its average is.

to Neutron

Можно показать, что существует однозначная связь, между показателем Херста и коэффициентом автокорреляции. Тут всё так же: <0 - тактика откатная, >0 - трендовая.

Hearst is less than zero? And what, curiously, is its relationship to the autocorrelation coefficient?????

 

You yourself are more than zero!

I was talking about autocorrelation coefficient r in the series of first difference of initial BP, it is true for it, not for Hearst: r<0 - tactics rollback, r>0 - trend. And the relation you are interested in, you can get on your own, examining diffusion coefficient for one-dimensional Brownian motion and relating it at first with Hurst exponent, and then with autocorrelation coefficient. Your qualification for this problem is enough!

 
surfer >> :
Is it fair to say that if a series of quotes has a Hyst value much lower than 0.5, then the tactic of opening positions against the outliers will be effective, assuming a high probability of coming back to the mean? And vice versa, if H is considerably higher than 0.5, then a trend tactic should be used?

>> Fair enough.

 
Neutron писал(а) >>

You yourself are more than zero!

I was talking about autocorrelation coefficient r in the series of first difference of initial BP, it is true for it, not for Hearst: r<0 - tactics rollback, r>0 - trend. As for the relation you are interested in, you can get it by yourself by considering the diffusion coefficient for one-dimensional Brownian motion and relating it at first with Hurst index, and then with autocorrelation coefficient. Your qualification for this problem is good enough!

https://ru.wikipedia.org/wiki/%D0%9A%D0%BE%D1%80%D1%80%D0%B5%D0%BB%D1%8F%D1%86%D0%B8%D1%8F

the correlation coefficient I know is a number.

autocorrelation coefficient is a function of time shift https://www.mql5.com/ru/code/8295

What is autocorrelation coefficient ? how is it calculated ?

WE will start to understand each other only when we clearly define the terms, give their precise and unambiguous definition in words and as a formula. If we don't do this, nothing will work. This is what happens month after month with searching for the mythical "trend" and "flat". Everyone has a different one, because there is no clear and unambiguous definition.

 
Neutron >> :

You're bigger than zero!!!

You're just flattering me! :о)))

I was talking about autocorrelation coefficient r in the series of first difference of initial BP, it is true for it, not for Hirst: r<0 - tactics rollback, r>0 - trend. And the relation you are interested in, you can get on your own, considering a diffusion coefficient for one-dimensional Brownian motion and relating it at first with Hurst exponent, and then with autocorrelation coefficient. Your qualification for this problem is enough.

And what's the connection to Hearst, you mathematician?

 
Prival писал(а) >>

https://ru.wikipedia.org/wiki/%D0%9A%D0%BE%D1%80%D1%80%D0%B5%D0%BB%D1%8F%D1%86%D0%B8%D1%8F

the correlation coefficient I know is a number

the autocorrelation coefficient is a function of the time shift https://www.mql5.com/ru/code/8295

What is autocorrelation coefficient ? how is it calculated ?

WE will start to understand each other only when we clearly define the terms, give their precise and unambiguous definition in words and as a formula. If we don't do this, nothing will work. This is what happens month after month with searching for the mythical "trend" and "flat". Everyone has his or her own trend, because there is no clear and unambiguous definition.

Sergey, look here (the uppermost post).

 
Neutron писал(а) >>

Sergei, have a look here (topmost post).

Looked at. It's 25 again. it's a correlogram, it's a function. A function turns into a number, only at a certain value of the argument.

"In time series analysis, a correlogram, also known as an autocorrelation plot, is a plot of the autocorrelations of a sample, from h (time lag). "

this is what it looks like 'Autocorrelation function' it is a graph !!!

Now what does the graph (function) get compared to a number ? so is it ?

Or maybe you just have to compare not a function but a number to a number.

Hearst index is a number and should be compared to a number !!!

Z.I. The correlogram and ACF are essentially a set of autocorrelation coefficients. It uses a single number "autocorrelation coefficient (one)". So I wanted to find out what it is, what do you think it is, at what value of the argument, the autocorrelation function becomes an autocorrelation coefficient. Some fix the ACF at 0.707, some through the integral - this is important for another problem. Determining the time interval during which a process is correlated with itself. (For traders, this is the time during which the observed process retains its motion characteristics).