Machine learning in trading: theory, models, practice and algo-trading - page 3562
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
Or maybe you didn't realise there's a much simpler way to do this. Honestly.
I really don't understand, why go from the general to the particular? You can always optimise a solution for a specific task, but here I am interested in discussing a tool that gives a universal solution for all variants of tasks.
again, not necessarily the first. Just K-1 points, and their indices modulo K are different.
Maybe. The point is that you get a linear system of equations for prices. I don't know if your condition is necessary and sufficient for the existence and uniqueness of its solution (I'm too lazy to think about it).
Yes, you get a slu. which, at first glance, always lacks K elements to solve.
But it has a very specific form - a ribbon matrix. Any twist/bend/straight line in MA simplifies things considerably. Oh, and additional conditions that the original - prices, in ranges, fluctuate and so on
so it's resolved - all the information is still there. This is both good and bad. Bad that the information stuck in MA will start to play the role of noise in NN/DL, when they think that HF is filtered, but it is still there.
PS/ Above the topic already enthusiastically recovered prices from MA by DL methods. Which in itself hints - everything is in place and there is a direct solution.
I really don't understand why go from the general to the particular?
maybe from complex to simple?
You can always optimise a solution for a specific task, but here I'm interested in discussing a tool that provides a universal solution for all variants of tasks.
Can't rounding be applied to any problem?
again, not necessarily the first. Just K-1 points, and their indices modulo K are different.
I was spinning as much as I could, because the statement "Mashka contains information about the series" was eventually replaced by the fact that it contains only itself :)
Twirled as much as I could, because the statement "Mashka contains information about a number of people" ended up only containing itself :)
first of all, with the unban :-) for how long...
and further - in our conditions MA-shka contains everything necessary...in general-abstract form it is difficult to calculate (when initial data for example go in one direction and without cycles, MA always increases and does not turn),
in our particularities everything is fine and everything is enough. But it is so, abstractly, you can say to the word had to "information disappearance"
maybe from complicated to simple?
Can't rounding be applied to any problem?
I can't support the conversation because I don't understand the point of the discussion.
What is the goal? You want to declare your new unique method - fine, write an article, conduct comparative tests, for example, on accuracy and performance.
I do not object to your active and productive activity.
first of all with the unban :-) for how long...
and further - in our conditions MA-shka contains everything necessary...in general-abstract form it is difficult to calculate (when initial data for example go in one direction and without cycles, MA always increases and does not turn),
in our particularities everything is fine and everything is enough. But it is so, abstractly, you can say to the word had to "information disappearance"
If I get banned, I'll make my own forum and write about the MoD there :) it's even better.
It's not the point... one thing is discussed, then the topic "floats" for some reason.maybe from complicated to simple?
Can't rounding be applied to any problem?
It can be even simpler: break the results of the decision function itself (model/probability outputs) into bins.
And apply threshold tuning/discretisation; for example, this is how they do it in cost-sensitive learning.
It is desirable to do calibration before that.
The definitions there are normal, not smoker's :)
If I get banned, I'll make my own forum and post about the MoD there :) that's even better.
Did you even sell the porno car?
You know what I like about you here, I have a friend who is one in the same, you're twins.
he bought an extremely rare toyota recently))))