From theory to practice - page 49
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
Yes, yes, of course...
Just tonight - need to get away from such a knockout... After all, I've already run the program on a demo account today, and it turns out all wrong there!
The reason - I still cannot understand - how exactly the tick data should be taken, so as not to destroy non-marking, to be able to use historical archives...
Schoolchildren from DC are quietly smoking in the background :)))
Schoolchildren from DC are quietly smoking :)))
Do you mean the relationship between increments? Or between prices?
And what do you mean by "two degrees of freedom"?
By degrees of freedom I mean the classical definition:
https://ru.wikipedia.org/wiki/Степени_свободы_(physics)
If the current price is connected to the previous price by a vector and the next price is connected to the current price by the same vector, then we finally have the notorious 2 degrees of freedom which completely describe the system. What is 2 degrees of freedom in statistics is approximately the same and in the market there simply MUST be a t2-distribution. And I can't find it... How so? I don't get it...
Yes, I accept all the rebukes.
But, it turns out that if there is no t2-distribution in the market, then it is pointless to use historical data.... I find this extremely frustrating...
There are a lot of smart and educated people on this forum
To get a profitable trading system is a hell of a job that takes years.
And not everyone succeeds.
Alexander, it only remains to wish you luck!
There are a lot of smart and educated people on this forum
Getting a profitable trading system is a hell of a job that takes years.
And not everyone is successful.
Alexander, I wish you luck!
Thank you!
Yes, I've got something, of course. But now I have to redo everything - Markov chains, of course, that's another story. And I may not make it to New Year's Eve...
Okay. I'll write if I find anything else interesting.
Good luck to all!
Sincerely,
Alexander and Schrodinger's cat :)))))))
By degrees of freedom I mean the classical definition:
https://ru.wikipedia.org/wiki/Степени_свободы_(physics)
It makes a very beautiful interconnected picture - if the current price is related to the previous price by a vector and the next price is related to the current price by the same vector, then we have the notorious 2 degrees of freedom which completely describe the system. What is 2 degrees of freedom in statistics is approximately the same and in the market there simply MUST be a t2-distribution. And I can't find it... How so? I don't get it...
Honestly, this is some kind of water. I'd ask you to figure it out and articulate clearly what you want to get/see, but you're ignoring most of the questions).
At the lowest level (if you don't go any deeper than ticks), there is a basic dependence of nth increments on (n-1)-th increments in the market. And it exists not only between n and (n-1), (n-1) and (n-2), but further back in history, gradually fading, i.e. there are a lot of "degrees of freedom" as you understand it. But there is no need to get bogged down in degrees of freedom.
Do you want to find something else in the market? Can you clearly articulate what? The dependence of what on what?
Without using the term "distribution", distribution does not describe dependence.
p.s. If you weren't so lazy, I'd show you in practice that the same bollinger works about the same with any way of reading ticks, and doesn't give a sneeze about the type of distribution anywhere.
Honestly, this is some kind of water. I would ask you to sort it out and articulate clearly what you want to get/see, but you ignore most of the questions).
At the lowest level (if you don't go any deeper than ticks), there is a basic dependence of nth increments on (n-1)-th increments in the market. And it exists not only between n and (n-1), (n-1) and (n-2), but further back in history, gradually fading, i.e. there are a lot of "degrees of freedom" as you understand it. But there is no need to get bogged down in degrees of freedom.
Do you want to find something else in the market? Can you clearly articulate what? The dependence of what on what?
Without using the term "distribution", distribution does not describe dependence.
p.s. If you weren't so lazy, I would show you in practice, that the same bollinger works approximately the same with any way of reading ticks, and it doesn't care about the type of distribution anywhere.
You have to prove this dependence, don't you? Right? So far I don't see that proof, although I VERY much want to see it. Then we could rightfully say - the process in the market is non-Markovian! and calmly use the database of archival data. Now we cannot say so with firm certainty.
Moreover, it can be considered proven from this thread - that when we read ticks at exponential time intervals with averaging, we have a MARKOVian process with no memory. There is no correlation between tick quotes in this case! That's certainly proven by practical results. No?
At the lowest level (if you don't go any deeper than ticks), there is a basic dependence of the nth increment on the (n-1)-th increment in the market. And it exists not only between n and (n-1), (n-1) and (n-2), but further back in history, gradually fading.
Here, by the way, is an excellent formulation of the main hypothesis of this thread. But, how do you know for sure that when reading data, for example, in 10 seconds, this dependence exists? Why not in 5 sec. and not in 1 sec.?
Actually, how exactly the quotes should be read, is not clear from this thread ... Alas...
You have to prove this correlation. Right? So far I don't see this proof, although I would VERY much like to see it. Then we could rightfully say that the market process is non-Markovian! Now we can't say that with any firm certainty.
So calculate the autocorrelation of the increments. Only this is unlikely to help the case, because dependencies of this kind are weak and impermanent.
Looking for patterns in prices is much more productive.
Suppose I were to tell you that the average volatility at a given time of day changes only slightly from one day to the next. Would that work as a "market memory"? And there are only two points in the prices, high and low, taken once a day. Tics are not even close with their increments.)
Wait a minute. Where has it been proven? ))) To prove the absence of memory, you have to rule out the existence of ALL possible (i.e. any) patterns. I don't recall where this has been done)
And again - a distribution of any kind contains no information about memory. Why have you suddenly decided the opposite?
And yes, I can easily build you an earning system on ticks with any readout method, even exponential, even completely random.