Machine learning in trading: theory, models, practice and algo-trading - page 1318
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
Here's the problem from Doc again
His model was able to solve the problem and made a profit on one of the artificial rows attached to that post. If your model can't do that - maybe it's too early to take on real BPs?
He was working with first differences - increments, and something else (he didn't tell me everything, of course). His model predicted the sign of the next increment and his signal was positive.
So why does everyone spit on the experience of successful traders and reinvent everything here, scaring the youth with some herbarium! I do not understand. I refuse to understand.
Alexander, how do you understand that this is the experience of successful traders? Kokos was riveting models on numeraire, his only monitoring was wiggling around zero. What kind of success are you writing about?
I just don't understand, where is the evidence base to assert something?
Here's the problem from Doc again
Here he had a model that could do the job and made a profit on one of the artificial series attached to that post. If your model can't do that - maybe it's too early to take on real BPs?
He was working with first differences - increments, and something else (he didn't tell me everything, of course). His model predicted the sign of the next increment and his signal was positive.
So why does everyone spit on the experience of successful traders and reinvent everything here, scaring the youth with some herbarium! I do not understand. I refuse to understand.
It is not known what happened to Doc, it is unlikely that he abruptly became successful and stopped contacting us.
People engaged in non-standard activities are in great need of communication, this includes businessmen and traders - they have no one to just talk to about what they devote most of their time to.
Basically, I don't work with increments at all here, I only work with predictors that describe the behavior of price in two-dimensional space. By the way, we cooperated with Doc, and he tested his script (which was supposed to select the best predictors for NS) on my predictors.
The idea of the herbarium is very simple, you probably just don't understand what we're talking about.
Alexander, how did you figure out that this is the experience of successful traders? Kokos was riveting models on numeraire, his only monitoring was wiggling around zero. What kind of success are you talking about?
From my personal correspondence it is absolutely clear to me that he found what he was looking for. I could be wrong - but he does not want to confirm or deny my words. He just won't get in touch. Well, let it be a forum legend :))
It is absolutely clear to me from my personal correspondence that he found what he was looking for. I could be wrong, but he does not want to confirm or deny my words. He just won't get in touch. Well, let it be a forum legend :))
Murphy's law - if you can be wrong, then you are wrong.
No legend, he did not demonstrate anything extraordinary, Alexey's pictures are not worse, as well as the scope of research, for exampleMurphy's law says if you can be wrong, you are wrong.
no legend, he did not demonstrate anything extraordinary, Alexey's pictures are not worse, as well as the scope of research, for exampleWell, okay - I'll shut up, I'll shut up.
Example: a task is set: here is a sinusoid with superimposed white noise, does the neural network predict such a thing? And so on and so forth.
Such a problem can be solved instantly with Fourier. Not knowing this for a physicist is like not knowing road signs for a driver.
Such a problem can be solved instantly with Fourier. Not knowing this for a physicist is like not having a clue about road signs for a driver.
Show the solution using a neural network. In tabular and graphical form.
Then solve Doc's problem (see above). Only after that it makes sense to argue about something.
How is it possible, not being able to solve such problems, to take up real BP - I do not understand. Some kind of herbaria to come up with... This literally blows my mind.
Show the solution using a neural network. In tabular and graphical form.
Then solve Doc's problem (see above). Only after that it makes sense to talk about something.
I've already posted it. This problem is so elementary that it doesn't deserve any attention. Almost every neural network package has examples with prediction of simple functions.
I've done 10s of experiments on graphs of different custom (composite) distributions as well. The more "Markovian" the worse, of course.
https://www.mql5.com/ru/forum/285454/page3
I've already posted it. This problem is so elementary that it does not deserve any attention. Almost any neural network package has examples with prediction of simple fs
I've done 10s of experiments on graphs of different custom (composite) distributions as well. The more "non-markness" the worse, of course.
https://www.mql5.com/ru/forum/285454/page3
Max, this post was not for you - I already know that you've worked with different functions - but for some kid.
Max, this post was not for you - I already know that you worked with different functions - but for some kid.
Well, just in case, that we are not just moving the language here :)