From theory to practice - page 51
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
And for the uneducated, I repeat - Bollinger Bands are good ONLY for Markovian processes and that's it. If you are convinced that they are not Markovian, why do you use them?
Yeah, she's gone now, Oleg... Disappeared last night...
as long as it's not at night and it's not from the real world.
but the memory is there, strangely enough
And you've been told for a long time - use the minutes (M1) in the MT5 terminalAs long as it's not at night and it's not from the real world.
but there's a memory.
By the way, right now my Expert Advisor is running at home and it is doing something stupid - with exponentially distributed time intervals, i.e. for a Markov process it takes into account the history and calculates something there... Interesting to see the results...
By the way, right now I have an EA running at home that does a silly thing - with exponentially distributed time intervals, i.e. for a Markov process it takes history into account and calculates something there... Interesting to see the results...
Yes, very
The memory MUST be! This is where I agree with everyone. And if my EA shows positive results, it will mean that there is not exactly a geometric distribution there. There is at least a little bit of a discrepancy.
Yes, once again, for the uneducated - the Bollinger Bands only take into account the CURRENT variance. You may have a great entry point. But, here's the exit point... Price does not have to and will not seek the average. It will tend if the entry point for the current variance is aligned with the entry point for the historical variance at a certain sample size. Read it again and remember it.
Where there is a geometric distribution, there is no memory and there cannot be one.
Once again:I propose an experiment. You lay out a series with a geometric distribution (whether real or generated), and I will show you how to add absolutely any regularities to it, i.e. "memory", without violating the distribution.
Yes, once again, for the uneducated - the Bollinger Bands only take into account the CURRENT variance. You may have a great entry point. But, here's the exit point... Price does not and will not tend towards the average. It will tend if current variance entry point is aligned with the historical variance entry point at a certain sample size. Reread it again and remember it.
First we draw (ourselves, mind you) the average. And then we declare that price to average (or average to price) will not tend. And what kind of average do you have?
Speaking of cats, including Schrodinger's. You don't like cats? - You just don't know how to cook them.
You need to switch to spherical horses urgently.
When I read the data in the sequence that I applied, I got a geometric probability distribution of increments, in fact, on what I was pointed out by esteemed Vladimir. This is the proof of the lack of memory of the process under given conditions.
This is NOT proof of a lack of memory in the process.
It only indicates that your method is inadequate.
Think about it: an objectively existing process has memory, and the process is not deprived of memory just because you have done some manipulation.
Once again:I propose an experiment. You lay out a series with a geometric distribution (whether real or generated), and I'll show you how, without breaking the distribution, to add absolutely any patterns to it, i.e. "memory".
I agree. The topic of distributions must be completed. Just a minute. I'll post the file now.