Machine learning in trading: theory, models, practice and algo-trading - page 1652

 
Igor Makanu:

- as a matter of principle

- I didn't bother with it.

- in the tester does not work

... well, as if only a shaman's tambourine for the complete set is not enough, although ... probably all start so - the other would be boring)))

Okay, I do not distract from the contemplation of what fascinates

I don't understand the irony, but what's wrong? What are you evaluating now? I didn't say it was a profitable trading strategy. I'm just showing a visualization of how a real neural network works. It works as it works... And yes, I like that it works at all. So what's the problem?
 
Evgeny Dyuka:
I do not understand the irony, what is wrong?

The point is that I emphasized the use of a strategy tester, but there are two opinions about the feasibility of testing:

- the tester lies

- things that work in the tester on the real account fail

But there is a third opinion - we should analyze why it happens, but usually nobody has enough diligence

 
Igor Makanu:

The point is that I emphasized the use of a strategy tester, but there are two opinions about the feasibility of testing:

- the tester lies

- things that work in the tester on the real account fail

there is truth in the third opinion - it is necessary to examine why it happens, but usually nobody has enough diligence

1. the tester is an excellent tool and it does not lie. He is needed to test the profitability of the strategy, that is his job. If there is no strategy and no profitability, there is nothing to test.
2. We should overcome the lack of concentration of consciousness. It is high time to understand that "profitable strategy" is utopia, it is a mirage, for which you can run all your life.

 
Evgeny Dyuka:

"profitable strategy" is utopia, it's a mirage that you can run after all your life.

Why?

 
Evgeny Dyuka:

It's high time to realize that "profitable strategy" is utopia, it's a mirage, for which you can run all your life.

I totally agree, that's why I'm doing the reverse task - trying to determine if the strategy has stopped working and another TS needs to be connected.

 
mytarmailS:

why?

Because no one has such a strategy. I don't mean a fantasy one, but one that makes more money out of money, and it's stable and on a regular basis.
 
Evgeny Dyuka:
Because no one has such a strategy. I don't mean a fantasy strategy, but one that makes more money out of money, and it's stable and on a regular basis.
No. There is such a strategy and it works all the time. Simply, you have to have a lot of money. Become a liquidity provider in a weak market and its main counterparty. Trade there against everyone else both ways and make consistent profits. Neural networks probably won't be useful - you need a different algorithm.
 
Evgeny Dyuka:
Because no one has such a strategy. I mean the one which makes more money from money and it is stable and permanent.

I'll tell you a little more, there is no strategy that does not make mistakes. Any profitable strategy must make mistakes. The question is: What is the price of its mistakes?

But I have another question. You keep posting pictures that you do not quite understand. So... I am half-heartedly following it. And my question is quite simple. What is the practical meaning of your research? How can you put what you are doing into practice? Many people just buy research forgetting why they do it..... So that's the question for you. So... sober you up...

 
Igor Makanu:

I totally agree, that's why I'm doing the reverse task - trying to determine that the strategy has stopped working and another TS needs to be connected.

I totally agree. There is an expression "do not marry the position" - just do not "marry" the TS. For any TS, there are market conditions in which it is unprofitable.

 
Mihail Marchukajtes:

I'll tell you a little more, there is no strategy that does not make mistakes. Any profitable strategy must make mistakes. The question is, what's the cost of making mistakes?

But I have another question for you. You're laying out pictures with not very clear drawings. So... I'm half-listening to it. And my question is quite simple. What is the practical meaning of your research? How can you put what you are doing into practice? Many people just buy research forgetting why they do it..... So that's the question for you. So... Sobering up ...

Good question. The practical sense is the same as for everyone else here - to make money. Only the way to do it is not through a war with windmills in the form of a "super strategy", but through the creation of a new product. Today neuro is not worse than many indicators, for instance, I would not recommend opening a position against its signals. The fact that this is not an illusion is confirmed by the API - the Expert Advisor is being used.
Reason: