Machine learning in trading: theory, models, practice and algo-trading - page 973

[Yuriy Asaulenko]

Maxim Dmitrievsky:

So it's not julia.h but julia.mqh :)

Directly unlikely, imho. More likely a DLL wrapper has to be created. MQL is not like C++, they are not interchangeable.

[Maxim Dmitrievsky]

Yuriy Asaulenko:

I don't think so, imho. More likely a DLL wrapper has to be created. MQL is not like C++, they are not interchangeable.

I'm leaving now, because I'm not able to do it :)

[Yuriy Asaulenko]

Maxim Dmitrievsky:

That's where I take my leave and go into the shadows, because I don't know how :)

I can't guarantee yet that I will do Julia, but I will look at her for sure).

[Maxim Dmitrievsky]

Yuriy Asaulenko:

I don't guarantee that I'll do Julia yet, but I'll take a closer look at it for sure).

The best way is to write entirely in Python or Julia, and the terminal will just trade

I don't have good tester, but if I will not use hft it will not be crucial. But later it will be easier to connect to other exchanges, when all mt5 brokers will buy my grail

[Yuriy Asaulenko]

Maxim Dmitrievsky:

The best way is to write entirely in Python or Julia, and the terminal will just trade

Of course, I won't have a good tester, but if it's not hft then it's not critical. But then it's easier to connect to other exchanges, when all mt5 brokers will buy my grail

Yep, everyone has to do his own job. The terminal - to trade, and the trader should make programs and use what he wants, not what he is given.

It seems that about half a year ago we were already talking about it).

I am sticking to this ideology since 2008. And I haven't cared which terminal for a long time. No preference.) And the speed in any terminal is enough - for scalping (by the cup and the strip) is enough with reserve, we are not dealing with HFT.)

As for the tester - it's just a cycle. It has a very small addition to the system, and the ticks can be removed.) If it is necessary.

[Maxim Dmitrievsky]

Yuriy Asaulenko:

By the way, there are testers for python with visualization and built-in example strategies

For Julia it seems that no

[Yuriy Asaulenko]

Maxim Dmitrievsky:

By the way, there are testers for python with visualization and built-in example strategies

Well, I haven't really worked with Python. I mean with modules - I haven't got the task for it yet. So far, everything is solved by the old tried and tested tools.

SZ Julia's packages - well, the most part of them are untested, and even non-working.( XGBoost Julia's package, by the way, is non-working, and that one is 2 years old. In general, judging by the dates, she, Julia, no one systematically engaged.

[Dr. Trader]

I agree:

Totally agree, that's why I stopped participating in that contest half a year ago :(

But now they finally did what they promised back in 2017 - reputation-based payouts. In the sliding window for the last 20 tournaments give a point for each live<0.693. The number of points is reputation, max=20. Current record in my opinion = 9. Participants with the highest reputation will be paid prize money even without stacking.

In short - those who know how to do forex - they have somehow learned how to make stable models, where the emphasis is not on 100% accuracy, but on the fact that in the future the model worked no worse than on trader. And now this knowledge can be monetized there.

[СанСаныч Фоменко]

Someone wants Python, so he uses any means to push his desire, passing it off as some kind of truth.

There are several ratings of programming languages, here in my blog I have twice cited such ratings.

Here's another one, one of the most authoritative TIOBE

Unlike the rating of R, in the rating of Python you need to understand a few nuances:

• there are many versions of this language, the two main ones are NOT compatible with each other, in search queries stackoverflow.com it is necessary to separate one version from another, moreover there are additional versions of this language, which must also be separated in search queries(Jython, Pypy, Iron-python, etc.
• When you use rankings, you should understand WHAT the ranking shows: google queries, education, job search, ... The main thing is the subject area. Python is more versatile than R and at the expense of outsiders to statistics it ranks above R.
• The competitors for R in the field of statistics and the corresponding visualization is SAS, which is far behind it.

[Vladimir Perervenko]

Not for discussion

https://www.kdnuggets.com/2018/05/poll-tools-analytics-data-science-machine-learning-results.html