Machine learning in trading: theory, models, practice and algo-trading - page 1414

 
https://gist.github.com/Roffild/08e289db362b3349858169c849c1918c- tried to calculate the significance of random forest predicates. The stats are fun, but I didn't find much sense in them. Dedicated to fans of the random forest...
Modification of a random forest with predictive significance calculations.
Modification of a random forest with predictive significance calculations.
  • gist.github.com
Modification of a random forest with predictive significance calculations. - ModifForest.mqh
 
Roffild:
https://gist.github.com/Roffild/08e289db362b3349858169c849c1918c- I tried to calculate the significance of random forest predicates. The stats are fun, but I didn't find much sense in them. Dedicated to fans of the random forest...
Apparently you calculated the frequency of use of each predictor. Didn't find anything in that either. Also, the list of importance is rearranged on restarts, just from the fact that the choice of predictors for the next split is chosen randomly.
 
Roffild:
https://gist.github.com/Roffild/08e289db362b3349858169c849c1918c- tried to calculate the significance of Random Forest predicates. The stats are fun, but I didn't find much point in them. Dedicated to fans of the random forest...

There are new technologies, CNN, LSTM on the clouds, there is no point in using Random Forest, it is not even decent to talk about it, we need to keep up with the times

 
Maxim Dmitrievsky:

Are you going to make us laugh in all the threads, you cardboard wisecracker?

What offended you now? This is the 21st century, no one writes scripts in asembler and syah, and in general all that could be written has long been written by professional scientists from Google and Microsoft, believe me that are much smarter and educated people like you. Normal people download the library or package and do everything you need in 5 minutes, not years of fiddling around like overgrown dependants.

 

That's it! Everything has already been invented and written. The development of civilization has stopped)))

Everyone switch to gadgets with widgets. Drop the widget with the button "Money", which is connected to something from Google or Melkosoft in 5 minutes

 
elibrarius:

That's it! Everything has already been invented and written. The development of civilization has stopped ))))

Everyone switch to gadgets with widgets. Throw the widget with the button "Money", which is connected to something from Google or Melkosoft in 5 minutes.

So it is, just many people do not understand it yet, the world is fully SPECIALIZED, everything is done by the professionals, that is, people who have all their life engaged in one thing, got education, diplomas, certificates, raised reputation in scientific circles, etc. Gone is the time when the big-aged dependents surfing the web, could pay attention to themselves that "dug" in the network that "interesting" as a little kid, like a cute puppy. In the next 5-10 years, with the introduction of total control and transparency, it will be generally forbidden to engage in non-profile activities after adulthood and this is correct.

 
Maxim Dmitrievsky:

You're a genius. Keep it up.

I'm not talking to you, I'm waiting for the code on Ruby

 
Maxim Dmitrievsky:

You remind me a lot of the stable boy in your epistolary style.

the same flight of fancy, weighing experience and intellect.

He's your teacher, no wonder, you admitted it yourself

Maxim Dmitrievsky:

If the other Teachers bowed their heads to the unforgettable Perepelkin, then I'll listen too.

I hope a new era will begin in our relations and a thaw

And so hypocritically begging, "hope", "thaw", "relationship"... Phew...

 
Aleksey Nikolayev:

Since this is now a branch on psychology, let me remind you of the Soviet psychologist Galperin (his death anniversary is today).

It seems to me that his theory of orientation to future action is quite suitable for describing trading.

https://ru.wikipedia.org/wiki/%D0%9A%D0%B0%D0%BD%D0%B5%D0%BC%D0%B0%D0%BD,_%D0%94%D0%B0%D0%BD%D0%B8%D0%B5%D0%BB

Happy accident[edit|edit code]

Associative mechanisms look for causes. A tendency tocausal thinking generates serious errors in assessing the randomness of truly random events. Kahneman gives the example of the sexes of six babies born in the hospital one after the other. The sequence of boys and girls is completely random: the events are independent, and the number of boys and girls born in the last hours has absolutely no effect on the sex of the next baby. However, it seems to people that the sequences M-M-M-D-D-D-D, D-D-D-D-D-D-D and M-D-M-M-D-D-M have different probabilities. Humans look for patterns everywhere and don't expect a random process to lead to regular results.


And of course, it doesn't change my attitude to psychology, it's just a set of things plucked from other fields, particularly from tervers

bestseller you can read it herehttps://avidreaders.ru/book/dumay-medlenno-reshay-bystro.html
 
Maxim Dmitrievsky:

My attitude to psychology does not change, just a set of things plucked from other fields, in particular from the theorist.

Believe me psychology is not a theorist, to compare them, and even more so to say that one came from the other in my opinion the top of cretinism.

But "happy accident", yes, that's yours, a reasonable choice in your case, adequate.