From theory to practice - page 888
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
WMA is probably not the worst choice - the equivalent of a FIR filter. But the choice of coefficients is a bit tricky. It's not so easy there).
It is hardly possible to come up with a better half-regression line, but the analysis interval should be relatively small.
And without rearrangement it is unlikely to work at all, either with FIR or with regression.
The floor regression works poorly on curves. it behaves normally only on smooth reversals.
Any equivalent MA with a longer period works exactly the same, but worse. There are no miracles.
By the way, reversals are not needed. You need an average for the theme, not a reaction to every sneeze.
You should get a near-normal distribution of linear deviations from this coveted average.
And how do we measure this normality.
For example, we measured the frequency of different deviations, plotted a bar chart in Excel.
how do we calculate "normality" now?
And how do we measure this normality.
For example, we measured the frequency of different deviations, plotted a bar chart in excel.
how do we calculate "normality" now?
Data --> Data analysis --> Descriptive statistics for a set of linear deviations from the selected moving average in the time window. This is how I did it.
the floor regression does not work well on bends. it behaves normally only on smooth pivots.
It remains to find a way to determine the "bend/turn" as quickly as possible and a good TS is ready :-)
while A_K is rubbing the excesses there are about two dozen potential grails, or at least good ideas :-)
While A_K is rubbing it out with excesses, there are about two dozen potential grails, or at least not bad ideas, thrown into the branch :-)
Oh, great.
Except that no one goes beyond ideas.
I remind you that the purpose of this thread is to provide practical results based on the chosen theory, and not just a dream or a drunken fantasy.
I assure you guys - a trading signal with a theoretical basis will have maximum credibility. It cannot be otherwise.
There seems to be a Nobel Prize offered for such a theoretically sound system, no? Just like a man who gets pregnant. I don't think anyone ever got rich.
:))) Funny. But agree - on a subconscious level, everyone wants to see that the signal is valid and not just for fun, as they call it. Shit, what can I say - just for the life of me, no one would ever accept any project if it didn't include references to literature.
Seriously - I have both the desire and ability to pay for any signal. But, without a minimum of theory - no way, I won't sign up for any.
:))) Funny. But agree - on a subconscious level, everyone wants to see that the signal is valid and not just for fun, as they call it. Shit, what can I say - just for the life of me, no one would ever accept any project if it didn't include references to literature.
Seriously - I have both the desire and ability to pay for any signal. But, without a minimum of theory - no way, I won't sign up for any.
I also have a desire, but having been disappointed by what I've seen, I started inventing my own systems )) Nothing as clever as a banal blind parametric optimisation has yet been invented
What if you take (Max per period + Min per period)/2 this value is not always in the centre of the channel?
I don't. That's why I want some noodle to do his research and present the results to the community.