Machine learning in trading: theory, models, practice and algo-trading - page 2190
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
You need a matstat - parameter estimation, hypothesis testing, etc. At least at the level of a clear understanding of what a p-value is, etc.
In econometrics, you can't do without it, either.
So it's hard to argue)))
These are of course overlapping directions, but they are fundamentally different. And preprocessing, smart adaptive filters with feedback, well it's.... And no one forbids these methods to process anything, VR or music. In ECM there are still matstat and probabilistic intricacies, which DSP does not have. Quantization of signal is good, but it always bothered me, periodical sampling of continuous signal in order to reduce size by something quantization...
In general, one does not interfere with the other, and if you know how to use it, it is good))) But ECM is not a part of DSP. They are different things. And DSP is part of mathematics, matstat and simply mathematics.)
And what do you not like about econometrics))))
and how are they fundamentally different?
i'm not saying i don't like it, i don't see any independence in this "science"
All that is in ECM is in DSP and at the same time DSP has a giant amount of other knowledge that is not in ECM.
To compare EKM with DSP and compare them to each other.
It's like comparing a book on ME with different algorithms, history, conversions, codes, types of ME and so on...
with some package manual which implements the same kind of neuronics...
NS will build any DSP for you practically for free
there is no need to discuss any filters. You need to use advanced neural networks to classify time series. That's all.
Actually ns handles any series better than other approaches. Except for the naive ones, where everything is clear as it is
NS will build any DSP for you practically for free
there is no need to discuss any filters. You need to use advanced neural networks to classify time series. That's all.
Better not any, but the right one)))
Better not any, but the right one)))
There are no necessary ones, only unnecessary ones.
How are they fundamentally different?
I'm not saying that I don't like it, I don't see any independence in this "science"
All that is in ECM is in DSP and at the same time DSP has a giant amount of other knowledge which is not in ECM.
To compare EKM with DSP and compare them to each other.
It's like comparing a book on ME with different algorithms, history, conversions, codes, types of ME and so on...
with a manual for some package which implements the same kind of neuronics...
ok, ok)) used to subscribe to the magazine Quant. Solved problems, and the concept of quantum for me has always been probabilistic. You know, with a certain probability of material and wave. And when quantum discrete mechanics appeared, well as if nothing, BUT when thinning began to be called quantization....))))))
NS will build any DSP for you practically for free
there is no need to discuss any filters. You need to use advanced neural networks to classify time series. That's all.
Actually ns handles any series better than other approaches. Except for the naive ones, where everything is clear as it is.
it's all from a deep misunderstanding))
BUT when thinning began to be called quantization....))))))
it seems to come from the west
Forum on trading, automated trading systems and trading strategies testing
FIR Filter with Minimal Phase
Oleg avtomat, 2013.06.05 08:21
Perhaps this is because econometrics has different goals, the main one is fogging.
;))
Forum on trading, automated trading systems and trading strategies testing
Econometrics: forecasting one step ahead
Oleg avtomat, 2012.01.20 10:13
Well... It turns out that there is a concept of band-pass filter in econometrics... And the authors of this invention are, of course, econometricians?
.... Advertising: I will calculate a bandpass filter of any complexity. Expensive.
:D))
Forum on trading, automated trading systems and trading strategies testing
Econometrics: forecasting one step ahead
Oleg avtomat, 2012.01.20 11:33
eh, faa... I should not read the package... but study the appropriate subject.
More and more I am convinced that all your econometric knowledge boils down to reading descriptions of several econometric packages... without understanding the what, how and why...