FIR filter with minimum phase - page 9
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
A filter is one way of smoothing economic data. If you look at the literature, the two filters that have the most use are Hodrick-Prescott and Kalman. However, there are a great many filters, all perfectly designed and applied in practice. Why such limitation of application of filters? The answer lies in a completely different area and has nothing to do with filters, phases or anything else.
If a filter has been applied to input data, the initial quote consists of two components: the result of filtering and some difference between the initial signal and the filtered result. Since in trading (unlike in radio electronics) we are always interested in the forecast, it is quite natural to ask: can we extend the filtering result forward? The answer lies not in filtering result, but in filtering residual (noise). If noise is stationary (mo and dispersion are approximately constant), we can extend filtering result, while dispersion will be an error of this prediction. If the residual is not stationary (has variable mo, which can be removed, and worse has variable and often very intricate variance), then the prediction is not possible, because the variance is about the past and has nothing to do with the future.
Conclusion: all talk about filter phase is meaningless, if we get non-stationary residuals as a result of filtering.
Klas is a good thread. Sorry, I will not insult electronics... It is economists who mess up our methods and call us names with their own words and think they are cool. ECONOMETRY is kind of science, while radio electronics are just idiots ... Just look around you. Radio engineers have made and are making what you all use here, computers, televisions, receivers, transmitters, mobile phones, microchips, processors, etc. etc.
What have the econometricians done? Where are their results? crises...bubbles...inflation...pyramids
Z.U. So excuse me, I have made forecasts all my life, there is the science of radar(who reads a lot of words in this thread). In a nutshell, I have to give (or rather an algorithm) to the weapon system where to shoot...where should the missile fired at the enemy aircraft, which is spinning like hell on a frying pan....
Well if you are an old man, then who am I?
Wow never thought I would be measured against a man for whom I have a bottle of cognac (soon to be a rarity)
here we go :-) but it's just as a friend
1. Privalych Registered on the forumin January 2006.
2. AlexeyRegistered in November 2006.
1:0
1. A military pensioner. 28 calendar years of service ... was born in the glorious city of Daugavpils 5.03.1967 ... VAU that to me already 46 it turns out
2. Sir, is it really 1:1, or else I have yellow trousers :-)
if i understand correctly this algorithm reduces group delay of the filter, i want to run it in matlabe
I will try to help the author of this thread. The answer may help. I hope it won't confuse you at least and won't lead you into theoretical mazes.
1. BIH and Kich filters are constant coefficient filters, which are selected, albeit scientifically, but once. At the design stage...I will try to give a clear example. You took two MAHs, ran them on history (found optimal values, according to some criterion, let it be the maximum profit) .... and now you put them on real.... hoping that the market will not change and we will earn money... so here you found some FIR filter, even if it has minimal group delay, ideal AFC, etc... etc.
But suddenly the market has changed and the filter hasn't, its coefficients are constant and .....
I'm not going to write further, the logical chain is clear.
Go into the field of adaptive filters, those that decompose movements into components, the most famous PF, decompose them... look at the spectrum... go ahead... until the spectrum changes you're in good shape. Example herehttps://www.mql5.com/ru/forum/127331/page3#348879 only don't forget that Fourier doesn't give a forecast, it tells you how the market is doing, sick or healthy. Other methods such as Kalman filter give prognosis. We compared it to Butterworth filter once...https://www.mql5.com/ru/forum/105740/page22#53541
Well, I'm almost fifty, I'll be exactly fifty in September (I'm about six years old in the picture). Nah, I've done the math exactly (haven't quite got round to figuring out my age by date of birth yet). But I'm not a military pensioner, I'm a civilian. And I don't have merits like you.
I meant that when I was talking about the old guys, not the registration period.
I remember about the cognac. Just at one time it seemed that you already despise me for being a moderator :)
Klas is a good branch. Sorry, I will not allow to offend radio electronics ... It's economists who have stolen our methods, and call us names with their own words and think they are cool. ECONOMETRY is kind of science, while radio electronics are just idiots ... Just look around you. Radio engineers have made and are making what you all use here, computers, televisions, receivers, transmitters, mobile phones, microchips, processors, etc. etc.
So what have econometricians done? Where are their results? Crises... bubbles... inflation... pyramids...
Z.U. So excuse me, I have made forecasts all my life, there is such a science as radar(who reads a lot of words in this thread). In a nutshell, I have to give (or rather an algorithm) to the weapon system where to shoot...where should the missile fired at the enemy aircraft, which is spinning like hell on a frying pan....
Why are you attacking an unresponsive child?
Econometrics is everywhere, from enterprise output planning to policy-based analysis and forecasting. Don't show ignorance, educated in another field man!
You want money in the financial markets, ditch all radar along with DSP and just enjoy TV (if you can).
The question of how to make what we are dealing with do what we need it to do is answered in cybernetics. Natural processes, whether physical or economic, are similar, hence the similar methods of analysis. The only difference is that in economics, there is much more room for chattering due to much larger characteristic time parameters of processes, so mathematical methods historically took root later. In radio electronics it is useless, there is a clear criterion - the device either works or not, while in economics, especially in macro, you can bullshit for decades, until everyone realises that it is bullshit. In the meantime, your fourth term as president has already come to an end, the profits have been earned, you can safely retire.
What a reply))
Klas is a good thread. I'm sorry, but I won't let you offend electronics... It's the economists who have taken our methods and called them names with their own words, and think they are cool. ECONOMETRY is kind of science, while radio electronics are just idiots ... Just look around you. Radio engineers have made and are making what you all use here, computers, televisions, receivers, transmitters, mobile phones, microchips, processors, etc. etc.
...