Dr.Fx corner - 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
I find it wrong to introduce additional error by adding up the squares of the deviations and extracting the root. For myself, I usually add up the moduli of the deviations. Sort of the same thing, but more primary. Thus, I estimate the degree of filtering by looking at the ratio of the sum of the moduli of the first differences in the output and in the filtered signal. See figures.
As if Kalman is better. But it is as if. Think about it: I have most of these twitches localized in the vicinity of the filter line. It goes back and forth, up and down, see? That is why strictly speaking the quality of the filter cannot be assessed that way. Volatility ratio is NOT a good measure of filter quality, because it does not take into account the nature of volatility.
... With just one SMA (you can also use your curve) you can build a dozen trading systems, and all of them will trade differently, some TS will be better and some will be worse.
I think it's wrong to introduce additional error by adding up the squares of the deviations and extracting the root. For myself, I usually add up the moduli of the deviations. Sort of the same thing, but more primary. Thus, I estimate the degree of filtering by looking at the ratio of the sum of the moduli of the first differences in the output and in the filtered signal. See figures.
What's the difference? It's the same thing. Here are the properties of a modulus. For a plane it is the root of the sum of the squares of the deviations (Pythagoras theorem).
Just add up all these deviations at all points and lay out the numbers. For all three filters. Three numbers and everything will be clear and straightforward at once.
How can you compare filters by adding up the differences (squares or moduli) of the deviations?
Then the best filter would be no filter - the difference in deviations is zero.
And in general, by changing my filter settings (it will no longer be the same filter as shown above to filter quotes, qualitatively the same, but quantitatively different), I can also get OEM (my operator name in matcad) lower than Kalman with the current settings - easily. If you set the task that way.
What's the difference? It's the same thing.
How can you compare filters by adding up the differences (squares or moduli) of the deviations?
Then the best filter would be no filter - the difference in deviations is zero.
That would be the worst filter. OEM (the name of the operator I have in matcad) will have the worst, maximum, value = 1.
Yes, I see. Mistakenly )))
PS then the filter line should be a horizontal straight line
And in general, by changing my filter settings (it will no longer be the same filter as shown above to filter quotes, qualitatively the same, but quantitatively different), I can also get OEM (my operator name in matcad) lower than Kalman with the current settings - easily. If you set the task that way.
you can make it even simpler to do nothing filtering, then the deviation will be zero.
I offered you a comparison methodology, we did that 8 years ago. You can use it, you can come up with your own criterion and compare as you like. You now have Kalman filter (the simplest version) as well as well-known Butterworth filter. We do not know your secret filter, so everything is in your hands. Most often a man is more confident in what he did and did it right, it helps a lot in the construction of TS, the main thing is to really try to do it right.
There is a saying. If you do everything well, it may be good, but if you do everything wrong, nothing good will come out of it.
This is not for you, but for many forex traders who have been building great theories for years. You are doing the right thing, look for the fish where the fish is. I sincerely wish you good luck...
Practice and time is the only criterion for the truth of all mathematical theories and tricks. And the path from high science to putting it into practice is thorny and long (usually).
You are doing the right thing, look for the fish really where they are. I sincerely wish you good luck...