Machine learning in trading: theory, models, practice and algo-trading - page 2012

New comment
 
Maxim Dmitrievsky:

The logic there is confusing as it is, I would like to understand it (which I have coded myself). And why does it work this way and not the other way?

Maybe it's the luck of the random logic? :)

 
Maxim Dmitrievsky:

Yes. It's adaptive control with reinforcement, not classical.

Are you going to describe the approach in Russian words in the article or here?

 
Aleksey Vyazmikin:

Do you plan to describe the essence of the approach in Russian words in an article or here?

No, I'm too lazy to write.

the gist of the approach has already been described

 
Maxim Dmitrievsky:

No, I'm too lazy to write.

the gist of the approach has already been laid out

I see.

With diving into the code, the details are revealed - and they are usually the point ;)

 
Maxim Dmitrievsky:

The cuts began to coincide +- with the tester... well, at least, if the real is not good, then in the tester too

For now it's trading since September 21, in the tester it's about the same. And for an earlier period in the tester it is good, but the weeks are unsuccessful (probably)

For some reason in the beginning of the week it is always good (after the "pre-training"), then it "drops", maybe just coincidence

found something else to improve, maybe it will finally work )

long fucking wait to catch bugs

on the long movements it pulls, on the U-turns stalls

here is an example of the tester (updated logic) the last 2 months

But it is worse with old logic. I will try to use new logic on a trader, I will freeze it next week.

From the 22nd trade till the last 127th is stomping on the same place, i.e. no gain

so it is a waste of time

 
Renat Akhtyamov:

from the 22nd deal to the last 127th is stomping on the same place, there is no gain

i.e. a waste of time

Fuck it, cap obviousness.
 
Maxim Dmitrievsky:
pshlnh, cap obviousness

Don't get mad at me, if it fails, I have nothing to do with it.

it's been said for a long time - the tester and the real time values are two absolutely incompatible charts

you should only trust your own logic and nothing else

I emphasize your logic, not the machine logic

We trust machine logic only if our own does not fail.

the algorithm is thought out and modified only by a man, and in the process of real trading and having previously thought about it with his own skull

If you rely on computer hardware, it is dead compared to the human brain.

;)

 
Renat Akhtyamov:

Don't get mad at me, if it doesn't work, then I have nothing to do with it.

it's been said for a long time - the state from the tester and the real deal are two absolutely incompatible charts

You should only trust your own logic and nothing else

I emphasize your logic, not the machine logic

We trust machine logic only if our own does not fail.

That is, the algorithm is invented and modified only by a man, and in the process of real trading and having thought about it with his own skull

;)

It's been said for a long time by various people - get out. No one even reads you.
 
Maxim Dmitrievsky:
It has long been said by various people - get out. No one even reads you.

There is a lot written here, but there is no result

Although I am wrong.

I have not seen anything but sinks

Not even a hint of profit from anyone
 
Renat Akhtyamov:

There's a lot written here, but there's no result.

although I'm wrong.

I have not seen anything but sinks

I would not be able to write enough. That's it, go away, do not pick up my phone
1...200520062007200820092010201120122013201420152016201720182019...3699
New comment