![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
... There are plenty of methods, tools, tests and experiences already developed.
... on self-organising stochastic processes with a random structure.
... In simple terms, there are several subsystems "competing" with each other. One subprocess takes over the control of forming a quote from the other.
The reverse side of this medal ("free competition") is a rigid hierarchical subordination.
In this case, it's not the "subprocess that takes control" (randomness), but the subprocess that takes control (goal setting). That's the twist.
ps. and this is not linguistic acrobatics ;)
to avtomat
но "где деньги, Зин?" (с)
Unfortunately, my colleague doesn't understand that these methods have limited applicability, and different formulations of the problem.
The other side of the coin ("free competition") is a rigid hierarchical subordination.
In this case, it is not the "subprocess that takes control" (randomness), but the subprocess that takes control (goal-setting). That's the twist.
ps. and this is not linguistic acrobatics ;)
yes, exactly so, that's what I wrote. This is the eleventh attempt to explain to a colleague, I am trying to pick up different words. I am experimenting.
to avtomat
My colleague, unfortunately, does not understand that these methods have limitations on applicability, and different formulation of the problem.
Yes, exactly so, I wrote about it. This is my eleventh attempt to explain to my colleague, I am trying to choose different words. I am experimenting.
I am discussing the verbal description of the model and there is no place for "intercepting control" "stochastic structure" etc. What we see is what we sing. We see a forward movement - we sing the trend, we see a level - we sing a breakthrough or a rebound. And you, this is a "perseptron". That is why I am deaf and blind.
So don't 'explain to a colleague', but strictly speak the same language. I have not yet started discussing the inner workings of your models.
I am discussing the verbal description of the model and there is no place for "overriding control" "stochastic structure" etc. What we see is what we sing. We see a forward movement - we sing the trend, we see a level - we sing a breakthrough or a rebound. And you, this is a "perseptron". So I am deaf and blind.
So don't "explain to a colleague", but strictly speak the same language. I have not yet started to discuss the inner workings of your models.
I wrote clearly and precisely (one of the variants of approaches):
The quoting process is reconstructed by mixing these two completely different processes according to certain rules. Here's an example https://www.mql5.com/ru/forum/134424/page33 along with the prediction. The classification only is different and so is the model. There are only two almost linear processes there. Everything else is the same.
It can't be that you didn't understand anything if you read it.
There is a separate field in statistics which studies, literally, "mixtures of stochastic processes". It assumes that a complex process (e.g. turbulence) consists of a superposition of many processes with different densities. Methods are being developed that decompose these processes. There's a lot of interesting stuff out there.
I've written clearly and precisely (one way of approaching it), quite simply:
The quoting process is reconstructed by mixing these two completely different processes according to certain rules. Here's an example https://www.mql5.com/ru/forum/134424/page33 along with the prediction. The classification only is different and so is the model. There are only two almost linear processes there. Everything else is the same.
It can't be that you didn't understand anything if you read it.
There is a separate field in statistics which studies, literally, "mixtures of stochastic processes". It assumes that a complex process (e.g. turbulence) consists of a superposition of many processes with different densities. Methods are being developed that decompose these processes. There's a lot of interesting stuff out there.
There are no classifications and long tails in economics.
There is production of goods and services that have inertia - trends
In markets there is a lot of goods - oversold and a lot of money - overbought --- levels
All this passes through the psyche of the crowd (which did not exist under socialism) and noise appears --- volatility
We trade trend, levels and volatility. You can also trade risk.
What do you mean?
There are no classifications and no long tails in the economy.
OK, so the long tails have already gone somewhere.
Then why did you talk about heteroskedasticity a bunch of times? (and yet ARCH processes are characterised by long tails)
So, the long tails have already gone somewhere.
Then why did you talk about heteroskedasticity a bunch of times? (but ARCH processes are characterized by long tails).
Nothing has gone anywhere. These are different levels of conversation. first there has to be a descriptive model, in words. It should be based on meaningful concepts. At the next level, the next formalisation is done, then some more. Heteroscedasticity is not a first level concept. And it doesn't have to be in the quotient at all. Kotir can be so transformed that there will be no heteroscedasticity.
You can't mix things up. And most importantly, one should not build applied models without a meaningful description. This is not general algebra.
Nothing is going anywhere. These are different levels of conversation. first there should be a descriptive model, in words. It should be based on meaningful concepts. At the next level, the next formalisation is done, then some more. Heteroscedasticity is not a first level concept. And it doesn't have to be in the quotient at all. Kotir can be so transformed that there will be no heteroscedasticity.
You can't mix things up. And most importantly, you can't build applied models without a meaningful description. This is not general algebra.
So that's how it is... Really?... So, all general algebra constructions don't make any practical sense? right?
:)))))
The other side of the coin ("free competition") is rigid hierarchical subordination.
In this case, it is not the "sub-process that takes control" (randomness), but the sub-process that takes control (goal setting). That's the twist.
ps. and this is not linguistic equivocation ;)
I'm just shyly asking: aren't goal-oriented systems a good idea at all?
... and - what is control: necessarily goal setting?
SZZ Sorry, it's not Friday.