Machine learning in trading: theory, models, practice and algo-trading - page 1658
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Opened the article on the advice of Igor Makanu
https://www.mql5.com/ru/articles/1530
Listed immediately to the "conclusions" and I liked one)
"Let's conclude this article with a very tough question: maybe the analytical part of the vast majority of TS is useless, and the main efforts should be focused on effective and sound money management methods ("Analytics is nothing, money management is everything else!")?"
I think the author of this article decided to write two articles to express this thesis between the lines ;)
I think that the author of this article seems to have written two articles in order to state this thesis in between the lines ; but the problem is deeper, you can see Savvateev on YouTube, not a series of lectures, but just his interview, he has expressed his opinion about the meaninglessness of time series forecasting and about the economy, and he is quite interesting as a man, no bombast, but he explains very well, in general I like Savvateev sometimes watch - I recommend it.
I'm not trying to measure myself with anyone, and I'm not surprising anyone! You're just sick...
At least one healthy person was added to our ward )))))
I updated the Expert.
This displays and combines two forecasts - short and long.
The long one is lighter and higher and the short one is darker and lower.
As candles are added the forecast is updated.
Download here.
If you zoom to a minimum will be about this picture.
I have updated the Expert.
This one displays and combines two forecasts - short and long.
The long one is lighter and higher and the short one is darker and lower.
As the candles are added the forecast is updated.
Download here.
If you zoom down to the minimum, it will be about this picture.
Zhenya, well, it's not interesting, honestly. In the first place because it is not clear, and in the second, there is no practical sense. Now make arrows where the forecast changes from buying to selling and vice versa. Your screenshots do not even allow you to assess the quality of the network.
What are you hoping for by displaying pictures of this kind?
Mihail Marchukajtes:
What are you hoping for by displaying pictures of this kind?
I hope that the picture turns you on and you write something duplicating the download link. From this forum is indexed well. As a result, my API is already falling down from the load. Do not skip pictures!
I hope the picture turns you on and you write something duplicating the download link. This makes the forum indexed well. As a result, my API is already dropping from the load. Do not skip pictures!
My algorithm...
Evra 5m, three days of trading from the evening of 26.03 to now, full OOS with no tweaks, no optimization of entry parameters, nothing, just as it is!
MSUA
Same thing, already trading random forest
Sorry, but you just admitted that you want to make money on indexing or whatever but not on trading. Your pictures are incomprehensible and make no sense. So do not bother yourself and the great admirers of this thread. Your network is rubbish and that's it. And if you have a resource from which you're going to promote your bullshit, be glad I'm not there, because I would have smashed it to smithereens. There's no point. There's no practical use!!!!
My algorithm...
Evra 5m, three days of trading from the evening of 26.03 to now, full OOS with no tweaks, no optimization of entry parameters, nothing, just as it is!
MSUA
Same thing, already trading random forest
Beautiful, if it's on a real account, it's super.
Nope, on paper so far, but pure OOS
My algorithm...
Evra 5m, three days of trading from the evening of 26.03 to now, full OOS with no tweaks, no optimization of entry parameters, nothing, just as it is!
MSUA
Same thing, already traded in a random forest
Immediately feels adequate approach, just remember that it may be a random OOS. Need non-repetitive optimizations with appropriate OOS, to make sure that the approach is adequate. All in all it is quite good. Unlike Eugene, who can not provide such a status...
The trick here is that you can't always be sure that the OOS will be gaining. Only multiple tests over time will answer that.