Machine learning in trading: theory, models, practice and algo-trading - page 1727
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
This picture shows the settings of the CSSA-cycle filter, which we find by cointegration when a perfect circle is obtained on the oscilloscope.
This picture shows CSSA-cycle filter settings that we find by cointegration when we get a perfect circle on the oscilloscope.
Take a look at the tool help there. It's just that if it were possible to optimize automatically and get this circle, this tool would be useful.
Is it even worth the time? I'll read it... but not right away.
It's the Caterpillar, and it doesn't work.
It finds a pair of loops and finds a lead-lag connection between them. But since these loops don't work, neither does everything else.
Well, Max, such a layout wrote and the site glitched. It can't be repeated now. What kind of crap is that?
But without trying not know, I say that CSSA, this caterpillar that does NOT redraw. Pis6tion, such a shame with the lost text :-(
I can hardly keep my mate :-(. The point is that the simple parameter optimization is not suitable, here the eigenvector tool that finds such parameters of the cycle settings when the cycle becomes anticipatory for the price or is ahead of it, or for some time becomes the reason for price changes, is just suitable. The caterpillar itself is not as important as its tuning parameters, when it is successfully tuned it can work a lot in the future and the circle on the oscillator's screen is the way that allows to do it theoretically. Anyway I had it once by accident with a proof in the not too distant future. That's why I'm sinking for this topic....
Without filtering it's a waste of time
Moreover, to decompose into components is not even a goal. Roughly speaking, to get a stable cycle you must transform the series in such a way that cycles are there from the beginning. Otherwise, further decomposition through eigenvectors (principal components method or caterpillar) will always show shit that changes with time.
I.e. this thing works only on pre-cyclic series
As forCSSA, I haven't found digestible information, what it's compared to.
I can hardly keep my mate :-(. The point is that the simple parameter optimization is not suitable, it is the eugene vector tool that finds such parameters of loop tuning when the loop becomes anticipatory for the price or is ahead of it, or for some time becomes a reason for price changes. The caterpillar itself is not as important as its tuning parameters, when it is successfully tuned it can work a lot in the future and the circle on the oscillator's screen is the way that allows to do it theoretically. Anyway I had it once by accident with a proof in the not too distant future. That's why I'm sinking for this topic....
Eugene Vector ))))
))))
Eugene Vector ))))
))))
Switch to python, we'll make a working environment for grails
switch to python, we'll make a working environment for grails
R is better))