Machine learning in trading: theory, models, practice and algo-trading - page 2989
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
The saddest thing is that you don't even realise what nonsense you're talking about....
It's not me, it's the speaker in the video. He's presenting everything, everything, everything he can think of. Watch it again if you don't remember. That was the idea.
And don't forget the second part of my post. In a case where there is a legitimate physical process (gas vortices) I admit it can help. But not in markets where there are no physical processes and lots of noise.It's not me, it's the speaker in the video. He's pitching anything and everything he can think of. Watch it again if you don't remember. That was the idea.
1) it feeds everything, and then the algorithm picks the best combination of features and even the models in the video...
How else do you find the best combinations? There's no other way.
You're gonna do it manually??? Or maybe you think you know better what features the models need, good luck with that.
2) The author solves real problems, and successfully solves them, and it is in the video.
3) The author made a comparison of his method with others, on different problems, and his algorithm on 60% of problems was the best for the best solutions even with a 30-40% difference, and this is in the video.
4) the author has been working on this method for a long time and is not alone, there is a team there and it is in the video.
And of course, of course the author is a sucker and has never heard of "rubbish in - rubbish out", as well as his team....
They certainly know and know how and have done less than you, haven't they? ;)
So you know better how to forecast BP than the team that develops probably the best BP forecasting library???
That makes sense, doesn't it?
1) it feeds everything everything everything and then the algorithm selects the best combination of features and even the models in the video.
How else do you find the best combinations? There's no other way.
You're gonna do it manually??? Or maybe you think you know better what features the models need. Good luck with that.
2) The author solves real problems, and successfully solves them, and it is in the video.
3) The author made a comparison of his method with others, on different problems and his algorithm on 60% of problems was the best for the best solutions even with a 30-40% difference, and this is in the video.
4) The author has been working on this method for a long time and is not alone, there is a team there and it is in the video.
And of course, of course the author is a sucker and never heard about "rubbish in".
Rubbish in - rubbish out", as well as his team....
They certainly know and know how and have done less than you, haven't they? ;)
So you know better how to forecast BP than the team that develops probably the best BP forecasting library???
That makes sense, doesn't it?
DSP is logical where there are patterns. A 5 mm roughness of the seam will give one set of signals from sensors, very similar signals will be received from hundreds of other similar irregularities. An irregularity of 10 mm will be different, which, for example, can be classified as critical. And they will all give approximately the same set of signals. And easily recognisable. Harmonic decomposition and other DSP techniques increase the accuracy of prediction for this physical process.
Tasks where their algorithm was better were probably also with patterns, not noise, as in the markets.
By the way, they worked on the task, sort of solved it, but it did not come to implementation. Apparently, the customer chose a better solution.
DSP is logical to apply where there are patterns.
Have we talked about DSP? Have I ever mentioned that word in a conversation with you?
We were talking about the fact that you have no competence to criticise the library, if you are surprised by the overshoot of features...
Tasks where their algorithm was better were probably also with patterns.
That's a great statement... which is based on a hunch, which is based on nothing, which is based on incompetence...
Tasks where their algorithm was better were probably also with patterns, not noise like in the markets.
Yeah, clearing noise is not about DSP ...
I'd rather be banned forever ...
Nice statement... which is based on a hunch, which is based on nothing, which is based on incompetence...
One would think you've reviewed and delved into all the problems they've solved..... so you contradict with guesswork too.
If their product is so great, why aren't they on Forbes lists (having made money on the market) but put it out for free because they don't give money for it ))))
The conversation is off topic. About the whistling in the pipe.
I'd rather be banned forever...
I'd rather be banned permanently...
Don't take everything but your loved ones personally))))))
By the way, they worked on the task, sort of solved it, but it didn't come to implementation. Apparently, thecustomer chose a better solution.
The customer most likely found a better solution in terms of money, not a better solution to the problem. i.e. stopped financing or sold the company).
Gas molecules are just less complex than humans, but the same enough, otherwise there would be no sociology, psychology))))
Don't take everything but your loved ones to heart)))))
I'm just allergic to stoners.
Seasonal is nothing. ML enthusiasts have year-round aggravation).
If they would do it in their own thread, without flooding this one, there would be no questions. By the way, it also concerns you.
I'm just allergic to stoners.
I'm afraid there are no others. Years of wasting time and no one's achieved anything of substance. And still. that's a sign of what? Everyone's pushing their own agenda that seems hopeful. That is why we do not agree with each other.
The customer most likely found a better solution in terms of money, not a better solution to the problem. i.e. stopped financing or sold the firm).
It is unlikely that they asked for an excessive amount of money for what was eventually put out into the public domain. Besides, it's a gas company. There's a lot of money in it. The best solutions can be chosen.