Machine learning in trading: theory, models, practice and algo-trading - page 2058

[Александр Алексеевич]

Now you open buy, then close buy and open sell, and during the flat you get minus trades, and when there are two exits during the flat, the network learns not to do anything

[Aleksey Vyazmikin]

Maxim Dmitrievsky:

I just sat down, I wrote a parser.

Here's the model, you need to feed 15 last increments into it, on each bar. The increments are counted as price minus 5-period oblique average. Send to the function double catboost_model(const double &features[]), which is in the inluder

If the signal is greater than 0.5 then buy, if less - sell. The time frame is 15 minutes.

I've been studying it since September 1 till now.

I'll just leave it here ))

How does the class differ from the one by Aliaksandr Hryshyn?

[Mikhail Mishanin]

Might be helpful https://habr.com/ru/post/525076/

[Maxim Dmitrievsky]

Aleksey Vyazmikin:

How is the class different from the one posted by Aliaksandr Hryshyn?

the model calculation function is copied as is, it's the same everywhere

[Rorschach]

Maxim Dmitrievsky:

If you add days, hours, etc. to the features, it does nothing:

bestTest = 0.4918224299

I'm throwing it right away, so nobody messes with it.

It's complicated. Ordinary systems show a pattern.

[Maxim Dmitrievsky]

Rorschach:

It's complicated. Ordinary systems show a pattern.

What ordinary systems

[Aleksey Vyazmikin]

Maxim Dmitrievsky:

The model calculation function is copied as is, it is the same everywhere

Copied from where? It would be good to copy the function for regression from there and for multiclassification too. And the possibility to build asymmetric trees has appeared, I want it too.

[Rorschach]

Maxim Dmitrievsky:

What are the usual systems

We buy on a certain day of the week and a certain hour with different SLs and TPs. For 10 years there are systems in the plus, where the profit is several times greater than the maximum drawdown.

I use the same system year by year and month by month. It is obvious that some combinations of parameters appear more often than others.

Question to experts: Dear experts, I want to "scientifically" prove the significance of the result. For this purpose the law of large numbers and 4*sko are used. Can I use it for TS with unequal SL and TP, and how many trades do I need?

[Maxim Dmitrievsky]

Rorschach:

We buy on a certain day of the week and a certain hour with different SLs and TPs. In 10 years, there are systems in the plus, where the profit is several times greater than the maximum drawdown.

This has not been detected. At the most they live a year.

[Rorschach]

Maxim Dmitrievsky:

no such thing has been detected. At the most, they live for a year.

In its pure form it is hardly interesting