Market etiquette or good manners in a minefield - page 75

 
HideYourRichess >> :

The main conclusion is that if H-volatility equals 2H then arbitrage on such a series is impossible (proven mathematically). Otherwise, arbitrage is possible. There are two strategies, depending on whether H-volatility is greater or less than 2H. These are the basics. Plus some remarks about patrons etc.

Is the magnitude of the H conditioned by something?

 
paralocus >> :

Is the H value due to something?

>> In what way?

 
HideYourRichess писал(а) >>

The beginning is not quite clear in my opinion. What makes you think it should be done this way. From the point of view of correspondence between trading algorithm and partitioning algorithm, it is desirable that the first point coincides with the first vertex.

In addition, the question of correct processing of "gaps" in the data remains open.

Regardless of the starting point, the uncertainty associated with the arbitrary choice of the starting point ends at the first extremum of the price series. Further, everything breaks down unambiguously.

In the construction I cited, the first Kagi count is arbitrary and is determined by the starting point on the quotient. You could, of course, get into an argument about the appropriateness of the correct entry, but then you have to stipulate the final target.

As for gaps, gaps and the like, I haven't considered that in this context. And I certainly don't deny the need for such quoter processing for MTS.

Once again: I claim that the analysis block that forms the Cagy partitioning count in MQL-e contains two if-blocks and two assignment operators. In the script it looks like three lines.

I get the impression that you don't have such an algorithm, namely the kagi algorithm is no more complex than the renko algorithm.

Let's deal with that first and then my understanding.

So I argue the same thing - namely the cagi algorithm is no more complex than the renko algorithm (and no less).

And let's deal with your understanding of my understanding first. You can't do without it!

Paralocus wrote >>
I have attached the listing and quotes. Parameter P is a series of Open(in this case GBPUSD minutes - available in the archive) Parameter m is the number of elementary base splits s in one kagi step

Paralocus , that's not how Kagi is built. There is only one parameter - H (splitting step). At the chosen step of division the value Hvol is defined, which characterizes the profitability of the instrument (or its arbitrage).

 
HideYourRichess

I mean, is H calculated (i.e. what does H depend on?) or is it an arbitrary value?

 
Neutron >> :

paralocus, that's not how Kagi is built. There is only one parameter - H (the splitting step). At the chosen step of division the value Hvol is defined, which characterizes the profitability of the instrument (or its arbitrage).

I need to understand the algorithm of splitting, and to understand what Hvol depends on, if it depends on anything.

 
Neutron >> :

In Kagi splitting, no matter where you start from - the uncertainty associated with the arbitrary choice of starting point ends at the first extremum of the price series. Further, everything breaks down unambiguously.

Have you ever wondered how it is that for renko the start strictly corresponds to the first partition, but for kaga it doesn't. Don't you see any inconsistency in that?

Neutron >> :

In the build I cited, the first Kagi count is arbitrary and is determined by the starting point on the kotier. You can, of course, argue about the appropriateness of the correct entry, but then you need to stipulate the final objective.

The final goal is simple - to bring the trading model and the model of dividing the series into conformity. Important for practice. In theory, of course, such trifles are not important.

By the way, according to anecdotal evidence mentioned only in passing in the dissertation, the approaches that the author used in practice are just that, with them it is desirable to define the origin correctly. Especially on large H's.

Neutron >> :

As for gaps, gaps, etc., that's not something I've considered in this context. And I certainly don't deny the need for such cotier processing for MTS.

Accounting for gaps is important when it comes to practical use.

Neutron >> :

Once again: I claim that the block of analysis forming the Kagy partitioning in MQL-e contains two if-blocks and two assignment operators. In the script it looks like three lines.

So I'm arguing the same thing - namely the kagi algorithm is no more complex than the renko algorithm (and no less).

Demonstrate, please, I will either criticise him or admit I was wrong.

Neutron >> :

And let's get your understanding of my understanding sorted out first though. Not without it!

Let's see if it's worth sorting out first, maybe it's your mistake and not mine.

 
paralocus >>:

HideYourRichess

Я в том смысле, что рассчитывается ли Н (т.е. от чего Н зависит?) или это величина произвольная?

The value can be anything, on a random series. But in practice it has to be justified, to find the most appropriate one.


On a random series, H-volatility does not depend on H, on a real series it does, sometimes insignificantly, sometimes enough to overlap the spread. This, by the way, is a big blow to the theory of market fractality. If fractality existed, then H-volatility would always be around 2H, with any H.

 
HideYourRichess >> :

The value can be anything, on a random series. But in practice it has to be justified, to find the most appropriate one.


On a random series, H-volatility does not depend on H, on a real series it does, sometimes insignificantly, sometimes enough to overlap the spread. This, by the way, is a big blow to the theory of market fractality. If fractality existed, H-volatility would always be around 2H, at any H.

Thank you! In that case, I have a candidate for a suitable H value.

 

HideYourRichess, you are confusing two values - H-volatility and the H split step and thus confusing a person! I didn't accidentally introduce two different designations.

 
Neutron >> :

HideYourRichess, you are confusing two values - H-volatility and the H split step and thus confusing a person! I didn't accidentally introduce two different designations.

I understood that H-volatility is an average (or root-mean-square) length of all distances from local extrema to points of kagi-partition made with step H. I think you explained it this way. My question is about the H itself - where does it come from?