Machine learning in trading: theory, models, practice and algo-trading - page 2608

[Aleksey Nikolayev]

secret #:
You can also put it in a different way. By using MoD, you will definitely not get stability. Because to exploit market processes, you need to know them.

That is, to start from studying the structure of the market and not from curvaphysics.

I am already aware of your valuable opinion that you have to constantly practise the practicalities. But if you suddenly want to add something substantial and meaningful, don't hold back.

[secret]

Aleksey Nikolayev #:

They say the profitability and capital intensity of this algorithm increase significantly in spring and autumn)

With all due respect, it seems that your goal in the market is not to make money, but to pump up your speech skills)

[secret]

Aleksey Nikolayev #:

But if you suddenly want to add something substantial and meaningful, don't hold back.

So much more practical - google market opening at 10.

[secret]

Aleksey Nikolayev #:

I'm already aware of your valuable opinion that you need to practice the practical-practical-practical practice all the time. But if you suddenly want to add something meaningful and meaningful, don't hold back.

Here's another example: now gas sales will go for roubles, through intermediaries most likely.

And after that, the seasonal strategy on usdrub relating to tax payments by exporters is likely to break down. And it can be meaningfully switched off until the next market structure change.

And using MO, you have no chance of understanding why the incurred system broke and why it worked at all.

[Maxim Dmitrievsky]

Aleksey Nikolayev #:

To be honest, the scheme as a whole looks very sophisticated and it is unlikely that someone from the outside will be able to say anything meaningful.

1) There is some association with boosting, when each next model tries to improve error of previous ones.

2) I prefer the approach when we do not try to obtain one complex model for all cases but make several simple ones that work according to the principle of trade/not trade. You can make two: "buy" and "sell". You could have four: "buy after an up move", "buy after a down move"," sell after a down move", "sell after an up move". Maybe, more variants can be invented) Then they can be combined somehow - all kinds of creative options are possible here too).

As far as I know, it should search for something that is consistent, but I don't know what it is. This is on the topic of automatically generated strategies. So far I give it a "C", maybe there will be some progress. An untried topic.

[Rorschach]

Can this be considered proof that price is not "random" but a sequence of trends and flotsam?

[Aleksey Nikolayev]

secret #:
Here's another example: now gas sales will go for roubles, through intermediaries most likely.

And after that, the seasonal strategy on usdrub related to tax payments by exporters is likely to break down. And it can be sensibly switched off until the next change in market structure.

And if the sale for usdrub doesn't go through, then you can meaningfully not turn it off? Sounds logical)

secret #:

And using MO, you have no chance at all of understanding why the incurred system broke down and why it worked.

Sounds like saying that using MO leads to brain atrophy)

[Aleksey Nikolayev]

Rorschach #:

Can this be considered as proof that the price is not "random" but a sequence of trends and flotsam?

Honestly, the picture is not very clear without explanations. And it is better to draw a heat map or density chart.

[mytarmailS]

Maxim Dmitrievsky #:
By design, it should look for something - you don't know what, but which is consistent. This is on the subject of automatically generated strategies. So far I give it a "C", maybe there will be some progress. It's an untried subject.

First give me a definition of a pattern, and then build a model that looks for a pattern... Otherwise I am looking for something I don't know what.

[Renat Akhtyamov]

mytarmailS #:
You should first define the patterns and then build a model in search of patterns... I don't know what I'm looking for

He has one neura that trades, the other determines whether it is a positive pattern or not.

you'd probably need to add the intellectual part, as it were.

that is the part of the chart that led to the opening of the trade and the result, the pattern itself

like you suggested, if you remember

on the whole, the approach is interesting