Is there a pattern to the chaos? Let's try to find it! Machine learning on the example of a specific sample. - page 14
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
Well, I took 4 std indicators with different periods and trained on them for the last 12 years, the previous 10 years test (to the left of the dotted line).
There it just falls on the change of the global trend (orange chart), but the TS somehow holds on.
then you can see what kind of trades he opens on the chart and where, so that you can roughly estimate the principle and start from there.
Nobody disputes that there is a chance to describe with a small set of predictors any regularity on history, even in this thread there was an example of using only two predictors for this purpose, and there are articles on the portal, where using one polynomial describes the history of more than 10 years on minute TF.
Randomise indicators and their settings, and then train them - yes, it is an option.
So you have lost hope for automatic strategy building, and you want to be in full control of the decision making process now?
Nobody disputes that there is a chance to describe with a small set of predictors any regularity on the history, even in this thread there was an example of using only two predictors for this purpose, and there are articles on the portal, where using one polynomial describes the history of more than 10 years on minute TF.
Randomise indicators and their settings, and then train them - yes, it is an option.
So you have given up hope for automatic strategy building and you want to be in full control of the decision making process now?
Nobody disputes that there is a chance to describe with a small set of predictors any regularity on the history, even in this thread there was an example of using only two predictors for this purpose, and there are articles on the portal, where using one polynomial describes the history of more than 10 years on minute TF.
Randomise indicators and their settings, and then train them - yes, it is an option.
So you have given up hope for automatic strategy assembly, and want to be in full control of the decision making process now?
I don't want anything in essence, just as an option to get more meaningful TCs and ground for thought
And what do you see as the advantage of a strategy that can be described in words?
Random enumeration of indicator sets from 10 is fine, but from 100 is the curse of dimensionality. Sets have to be logically constructed somehow, randomly is not enough power.
Random is a matter of luck, otherwise it would be just an overshoot.
In fact, there is a problem of group learning, when a set of predictors is sampled for one phenomenon, but the model grabs something the best at the moment on the current sample. Most of the predictors I have are exactly like this. Ideally, I should build a tree on one group of predictors, then the next on another, but I don't know how to do that.
And what is the advantage of a strategy that can be described in words?
It's a matter of whether it's better to poke around randomly or stick to a priori reliable information
Information doesn't become more reliable by understanding it.
You will just know about a statistical phenomenon, but the cause of it will remain a mystery, most likely.
Information does not become more credible from understanding it.
You will just know about a statistical phenomenon, but its cause will remain a secret, most likely.
The branch question is certainly an interesting one....
That's why I was wondering.
Probably a regularity can be determined.
I suggest analysing several bars in a row, for example 3-4.
Then move one bar from the beginning of this sample of 3-4 bars and analyse again.
As if overlaying one sample on another.
It is possible to find a pattern
Like this:
As a theorist, tell me if you can train on a sample, then move the sample by one bar, train again, then move by one bar again and train again
and then overlay the results of the training on top of each other and see what happens.
So would the chaos disappear?
get the idea?