Machine learning in trading: theory, models, practice and algo-trading - page 940
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
That is, they change regardless of history, i.e. Q1 2016 is not like Q1 2017?
And fractals, so I have almost a fractal system for measuring price fluctuations in the range of 1 hour, 4 hours, 1 day, 1 week, 1 month. The planned fluctuation scale is calculated and we look where the price is at the moment(at what level).
it's not fractal))
It's just that every quarter the patterns change, sometimes drastically.
i.e. there is a high probability that the system will break down at the jointsit's not fractal ))
Just every new quarter the patterns change, sometimes drastically
How not fractal, small TF in my system is similar to large, or vice versa as you like, it is a fractal. But the frequency of similarities is not known, because it is defined by the function.
I.e. there is high probability of the system breaking down at the jointsI see, but fundamentally it should work :) Probably the quarter is one trend movement, and when this movement changes the system breaks down...
I see, but fundamentally it should work :) Probably the quarter is one trend movement, and when this movement changes the system breaks down...
Fundamental reasons - expirations, etc. Annuals are reports, as a rule.
Where there is a transfer of cash, there is always a change in the conjuncture.
I'd like to play with a search for cycles and google them. That would know when to train correctly, from what dates and to what
The main reason for the absence of stationary regularities is the constant change of capitalization and capital flows. Big capital flows rarely and slowly.
Has anyone been able to achieve an error of 0.2 or 0.3 on the OOS? Moreover, it often works on OOS
But the difference of 2-2.5 times as much as trayne strains me
I can not figure out when to finish development and begin in practice))
In Vladimir's articles.
What architecture? Can you give me a link?
On what architecture? Can you give me a link?
Well you've read it all...
Fundamental reasons - expirations, etc. Annual reports, as a rule.
Where there is a transfer of cash, the conjuncture is always changing.
I would like to play with a search for cycles, google it. That would know when to train correctly, from what dates and to what
The main reason for the absence of stationary regularities is the constant change of capitalization and capital flows. Big capital flows rarely and slowly.
Then it turns out that we have to increase the amount of data to search for such cycles, and do not sample for a year, but for 2-3 years and add numbers of months...
Both on Darch and on Elm - https://www.mql5.com/ru/users/vlad1949/publications - from the 4th article onwards the results in Acc are about 70% and higher.
Well you all read it...
not bad:
Yes, I read it, but diagonally, because I don't want to use R, it's not that much of a sporting interest.)
Then it turns out that you have to increase the amount of data to search for such cycles, and do not sample for a year, but for 2-3 years, add numbers of months...
I'm not sure, there's not much information on this.
But it turns out, for example, that if I sample the last quarter of the year, it works fine for the whole year, and then it breaks down...
something like that...
if short-term, it works for about 3 months, and then it breaks down ... i.e. again we get into a cycle, but quarterly