From theory to practice - page 1931

 
Grigori.S.B:

At the end of the stage?

I remember several times 2 years ago you were already 100% sure that the stage was not final, but already completed. And now it is most likely at least self-deception.

You may be a really good physicist, but a total zero in programming (which is strange). You are being offered help with elementary things that you can't handle. This help would advance the development a lot or mark off a dead-end branch, but you are afraid of losing the grail, which exists only in your imagination. Funny and sad.

:))) You're right. I don't know how to do anything. I'm just a total dummy :)))

Nevertheless, I don't need any help. I suggest you ignore my insignificant personality. Better start an interesting thread, because I write here purely out of boredom...

 
Alexander_K2:

:))) You're right. I don't know how to do anything. I'm just a total dummy :)))

Nevertheless, I don't need any help. I suggest you ignore my insignificant personality. Better start an interesting thread, because I write here purely out of boredom...


I wanted to e-mail you a picture, but I can't write it...

 
Evgeniy Chumakov:


I wanted to send a picture to the PM, but I can't write...

:))) Communication in PMs is closed forever, alas...

First of all I'm interested in time interval distributions between quotes. Preferably for at least 28 pairs on tick data. To compare them and to answer myself - why are they like that and how they should be. Identify the correlation with the time. Then - thin them out until they take a certain shape.

This is the Warrior's way and it has to be done by getting the attention of listeners and helpers. Not interested in PM.

 
Alexander_K2:


Then thin them out until they look a certain way.


Without experiments, I assume that for each hour of the day (and maybe even depending on the day of the week) the speed of price readout is different.

If you are talking about an exponential time window, the intervals between readings increase (or rather decrease) from the beginning of the day and towards the end of the day according to the exponential.

I could be wrong.

 
Evgeniy Chumakov:


Without experiments I assume that for each hour of the day (and maybe also depending on the day of the week) the speed of price readout is different.

If you are talking about an exponential time window, then the intervals between readings increase (or rather decrease) from the beginning of the day and towards the end of the day according to the exponential.

I could be wrong.

Generally speaking, this is correct. But, it's a very long way to reach a real, practical result - the Palma flow.

 
Alexander_K2:

By and large, true. But it's a long way to reach a real, practical result - the Palma flow.


I read something on Smart and it said something about a range, if the moving average is within that range, then the series is recoverable, and if it is, then it is not.

I think it may work as a trend/fleet key, but I don't know what to count there. I wanted to ask some advice here, but now I can't find this post.

 
Evgeniy Chumakov:


I read something on the Smart, it said something about a range, if the moving average stays within that range, then the series has reversion, if it does, then it doesn't.

I thought it might work as a trend/fleet key, but I still don't know what to count. I wanted to ask some advice here, but now I can't find this post.

No way.

Zhenya, there's no pattern to it.

If there was, everyone would have known the pivot point long ago

Mathematically I've worked it out, but still with a lag

the tester gave me the best result of all, but still stomping about

the trouble is the signal lag
 
Thinned out the tics today, probably not what's needed in the end.
Files:
 
Renat Akhtyamov:

No way.

Zhenya, there's no pattern to it.

If there was, everyone would have known the pivot point long ago

mathematically I calculated it, but still with a lag

the tester gave me the best result of all, but still stomping out of place

the problem is the signal lag

Or vice versa - ahead of the signal (if you compare at the time of a probable reversal

the corresponding ticks of several consecutive sales and purchases)?

 

On the subject of teak thinning... I think it's a tired topic in this thread. It's time to come up with something new.

One option would be to glue the candles together. It sounds mysterious and yet somehow simple and understandable.