Machine learning in trading: theory, models, practice and algo-trading - page 1653

 
Evgeny Dyuka:
Because no one has such a strategy. I mean not a fantasy one, but one that makes more money out of money, and it is stable and on a regular basis.

Well, you have an indicator, which predicts the market very well, right?

You realize that you're not the best expert of AMO in the world, right? And you're not the best expert on market mechanics, which means that you haven't exactly used the best solutions in your indicator. And now imagine that your indicator can be improved by 30-50% by applying the best solutions .... Can youimagine? )

 
Evgeny Dyuka:
It is a good question. The practical meaning is the same as for everyone here - to make money. Only the way to do it is not through the war with windmills in the form of a "super strategy", but through the creation of a new product. At present 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.

Exactly what I do not see in your pictures is specific signals. The construction of the predictive line above or below the kotir is not a signal. A signal is a specific condition, a competitive moment. Where are they marked in this way?

 
I don't trust the signals you use and I don't trust your company,Evgeny Dyuka:
Good question. I have the same practical sense as everyone else here - to make money. Only the way to do it is not through the war with windmills in the form of "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.

Do I understand correctly? That you don't want to trade but want to make money on the product itself?

If I understand correctly, it turns out that you do not believe in your product, do not trust the signals

 
Evgeny Dyuka:

I have not wasted my day, I have merged two forecasts in one on one chart - long and short.
This is the bit on the M1 for the last 24 hours. Colors:
1 - long down.
2 - long up
3 - short down
4 - short up

I can't tear myself away... It's mesmerizing. Not the fact that the forecasts are correct, just beautiful.


It is a pity I can not share the admiration, it is not clear where to sell, where to buy, where to stop. I wish the arrows...

 
Evgeny Dyuka:
Because no one has such a strategy. I mean not a fantasy one, but the one that makes more money out of money, and it's stable and permanent.

Yes :)))

 
mytarmailS:

Well, you have an indicator that predicts the market pretty well, right?

You realize that you're not the best expert on AMO in the world, right? And you're not the best expert on market mechanics, which means that you haven't exactly used the best solutions in your indicator. And now imagine that your indicator can be improved by 30-50% by applying the best solutions .... Can you imagine? )

That's right unless it's a random hit in the center of the target. Maybe there is no 30-50% improvement. This result is obtained by trying a huge number of options, there is no complete understanding of how and why it turned out what it did. I would have given the community the whole scheme if I fully understood where to go next, and now it is rather a random finding, so I do not disclose the scheme of selection of input data.
 
mytarmailS:

Well, you have an indicator that predicts the market pretty well, right?

You realize that you're not the best expert on AMO in the world, right? And you're not the best expert on market mechanics, which means that you haven't exactly used the best solutions in your indicator. And now imagine that your indicator can be improved by 30-50% by applying the best solutions .... Can you imagine? )

Such problems are not solved at the amateur level. I wouldn't say no to a team of two good high-paying programmers right now, one of whom is a mathematician with deep knowledge of neural networks, and the other is just a generalist programmer. Plus unlimited hardware costs. Under those conditions, you can talk about a result. It is possible only if there is a strategic investor.
Then there will be buy/sell arrows.
 
Evgeny Dyuka:
Such problems are not solved at the amateur level. Now I would not refuse a team of two good highly paid programmers, one of which is a mathematician with deep knowledge of neural networks, and the other is just a generalist programmer. Plus unlimited hardware costs. Under those conditions, you can talk about a result. It is possible only if there is a strategic investor.
Then there will be buy/sell arrows.
How sly you are. Would you happen to have a TS ready? Anyway, I see that there will be no practical application... Sadness...
 
Evgeny Dyuka:
Such problems can not be solved at the amateur level. Now I would not refuse a team of two good well-paid programmers, one of which is a mathematician with deep knowledge of neural networks, and the other is just a programmer of wide profile. Plus unlimited hardware costs. Under those conditions, you can talk about a result. It is possible only if there is a strategic investor.

That's not how it works, you have to become an expert in many fields yourself.


Highly paid specialists most likely already have a favorite occupation and a good salary. If they suddenly want to unite a mathematician neural networker and programmer to predict the market, why do they need you?

And about the investor, the investor is a man who wants to invest a dollar and get two dollars for a certain period of time. What do you say to him? You: We are like three tough guys who want to try to predict the market, we do not know if and when we will succeed, but we need the money now! )), the reaction of the investor?

 

Finally I made a crumpled output to the TA program from my AMO.

I say right away it's just a rough draft, which is deprived of much, for example without adaptive adjustment, which should be on every bar (I have not yet implemented it in the output), that is, the algorithm trades yesterday's morning knowledge of the market, and this is certainly not optimal, but we will finish it.


The algorithm should also be considered as the first block of three:

block 1 : market direction

block 2 : best entry point

Block 3 : filtering false entries.


That is the entry points are not clear not nice, but it is not the task of this block, the task of the block to be adequate to the direction of the market

Trading is essentially without parameters, one logical rule for entering a position, and that's it.


This is the Euro 5 min. if anything) (the most unpredictable and everyone's favorite instrument)


I wonder what will happen to him by the end of the day, because he's already using not actual knowledge about the market ... Let's see...

It's Friday, there should be sharp, strong movements, it will be a stress test.