FIR filters - page 6
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
Well said.
There is, by the way, a good filter, JMA, but even it cannot be traded. Although there are several reviews on the site of the author, which kind of implies how great it is to trade with him. But I don't want to call him a scammer at all.
What do you say, what are your hypotheses about its algorithm?
For me it is better to give the chart as a base for making a trustworthy statement.
i don't know about zhumurik, i had no need to investigate it. let them sell it to sloths. at least it's kind of adaptive.
I don't get it. My question is as follows: now we have taken a spectrum and identified its extremes. Can we use this information only to build filters that would provide the maximum profit at the given section? (in case of a primitive "flip" strategy).
in this thread I would prefer to discuss technical points rather than strategies...
e.g. how to use certain generator options or filter options from Behemoth
Then my next question is: Here's a 500 bar spectrum of the H4 EURUSD for example. Based on it, what would be the optimal parameters for satl/fatl? (P1,D1,A1, beats)
I'm interested to understand if this dependence is formalizable and if this process can be automated
fattles/satles is sort of a term referring to a specific Kravchuk strategy, whose articles can be found online, e.g. http://tauraspartner. nm.ru/metod.html
P1 is the main resonant wave of the LF filter
For example, there is a significant peak at 55 bars and you should set one of the filters of the strategy on it i.e. P1=55
D1 can be set inversely to the nearest dip of approximately 45~50 bars
The shifts are better to be minimum 0.0001 dB
and then within the acceptable order of the filter choose attenuation, for example A1=60 dB
let's look for other peaks and set other filters in the same way (if a strategy has several filters)
And as for automation - think for yourself which algorithm to code on µl, the library DF.dll is attached to help
Before relying on someone else's programme. At least check it at test points. This is a black box for many people. At least try to get what is shown in the picture above, which is called such a big word as spectrum. It's not a spectrum drawn there, at best it's what's left of the spectrum.
The spectrum looks very different !!!
Let them call it whatever they want.
the instantaneous spectrum looks like vertical lines
but it's useless to calculate a matched filter
Before relying on someone else's programme. At least check it at test points. This is a black box for many people. At least try to get what is shown in the picture above, which is called by such a big word as spectrum. It's not a spectrum drawn there, at best it's what's left of the spectrum.
The spectrum looks totally different!
You're not right at all, professor. Damn! Words are not enough to express how wrong you are :o) Everything is absolutely normal with the author of this prog. The very real spectrum. Classic!!!! They teach it in institutes.
fattles/satles is sort of a term referring to a specific Kravchuk strategy, whose articles can be found online, e.g. http://tauraspartner. nm.ru/metod.html
P1 is the main resonant wave of the LF filter
For example, there is a significant peak at 55 bars and you should set one of the filters of the strategy on it i.e. P1=55
D1 can be set inversely to the nearest dip of approximately 45~50 bars
The shifts are better to be minimum 0.0001 dB
and then within the acceptable order of the filter choose attenuation, for example A1=60 dB
let's look for other peaks and set other filters in the same way (if a strategy has several filters)
About automation - think yourself which algorithm to use on µl, library DF.dll will help you
Thank you, something is starting to become clear! I'll deal with the library, but now I need to deal with the calculation of the spectrum. The spectrum in our case, if I'm not mistaken, is the amplitude-period relation. First of all it is not clear, in what units is amplitude measured in GCM? Secondly, I now took a spectrum with the "Spectrum Analyzer" software from finware, which I've heard positive feedback about, and the results are completely different, although the algorithm is the same (MESA).
You're not right at all, proffesor. Damn! Words are not enough to express how wrong you are :o) Everything is absolutely normal with the author of the prog. The very real spectrum. Classic!!!! They teach it in institutes.
It might be a spectrum, only it doesn't have periodic components that are worthy of attention.
Please draw amplitude in logarithmic scale. Frequency preferably too.
And don't tell me that 3dB signal/noise is enough to claim anything when using FFT.
The calculation error is much larger.
You're not right at all, proffesor. Damn! Words are not enough to express how wrong you are :o) Everything is absolutely normal with the author of the prog. The very real spectrum. Classic!!!! They teach it in institutes.
Try to get at least a "C" in the subject. Only a man who has never built a spectrum himself can say that.