Discussion of article "Introduction to the Empirical Mode Decomposition Method" - page 4

New comment
 

This is one of those good and useful articles. Thank you so much, i was so blinded that my only way of analysis was on direct data stream. This is something that could prove really useful to me along with Heuristic Analysis. I will make my own implementation in the future and try to use the embedded MT5 charting functions instead.

However, i'm running the script with the graphical tool provided, it compiles and displays but data is shown as a square wave pattern only. Is this a normal behaviour?
I tried different symbols at different time periods on different date intervals, also i tried to set the 'n' value to something smaller then larger but attain the same results.

EMD Square Wave Data

 
Actually, this is not a good article. EMD is not a causal technique. This means that its past values change in real time, rendering it utterly and completely useless for trading. It's in the same category as Singular Spectrum Analysis, the Hodrick-Prescott filter and all types of splines. It looks o so good on a static chart, but in real time it is no better than a LWMA. Just place a SMA(1) on the result of your EMD line and you will see how bumpy it becomes... Nice from a research/scientific point of view, useless in trading.
 
MisterH:
Actually, this is not a good article. EMD is not a causal technique. This means that its past values change in real time, rendering it utterly and completely useless for trading. It's in the same category as Singular Spectrum Analysis, the Hodrick-Prescott filter and all types of splines. It looks o so good on a static chart, but in real time it is no better than a LWMA. Just place a SMA(1) on the result of your EMD line and you will see how bumpy it becomes... Nice from a research/scientific point of view, useless in trading.
Thanks for your feedback, it seems that it has been more useful than the article itself. I've been studing what you said and although this article has opened my eyes to other types of analysis, if i implement it most probably it will be as you said: BUMPY. Will spend my time on something else and i may use this for research purposes not for trading.
 

1. https://en.wikipedia.org/wiki/Hilbert%E2%80%93Huang_transform

2. And Google picture of Empirical Mode Decomposition. 

3. This probably one of my dumb comment of so many :). A little bit irony in here. The first thing to do, before calculating EMD, is to find the maxima and minima (see below). If we could do that already, then we're already make money. Finding maxima/minima is what we do around here. 

The wikipedia also mention (under limitation) that "Datig and Schlurmann [2004] did the most comprehensive studies on the performance and limitations of HHT with particular applications to irregular waves. ... The authors discussed using additional points, both forward and backward, to determine better envelopes.".  

4. Filtering out noises - that's what this all about. 

 

Hilbert–Huang transform - Wikipedia, the free encyclopedia
  • en.wikipedia.org
The Hilbert–Huang transform (HHT) is a way to decompose a signal into so-called intrinsic mode functions (IMF), and obtain instantaneous frequency data. It is designed to work well for data that is nonstationary and nonlinear. In contrast to other common transforms like the Fourier transform, the HHT is more like an algorithm (an empirical...
 

Thanks for the article, I enjoyed reading it.

A few points relevant for any decomposition (not only EMD):

  1. In Figure 6, the smoothing, as I understand it, is done through a single application of EMD to a given interval. It is still more correct to show smoothing through shifting a fixed-size window and applying EMD to each of them. Getting the smoothing value at the right end.
  2. In this sense, it is very interesting to compare the results of smoothing by right end and left end values (do the same for all smoothing functions).
  3. The forecasting of IMF functions seems to be as successful as the forecasting of simple MA. Since the basis of forecasting is the same - choice of extrapolator. Can you share a little bit?
    A detailed analysis of forecasting methods built on the basis of эмпирической модовой декомпозиции is not given here, as this topic is beyond the scope of this article.
  4. I like the idea of detrending very much, thanks.
  5. Applying EMD to price BPs clearly requires some transformation of the BPs themselves. Correspondingly, the same applies to the algorithm for finding extrema.

On EMD:

  1. Does the EMD result strongly depend on the type of spline function used?
  2. Comment on the reason for choosing such a construction criterion in EMD:
    At any point in the empirical mode, the mean of the envelopes defined by local maxima and local minima must be zero.


P.S. For comparison of smoothing functions it is worth, of course, to develop once and for all a criterion of smoothing efficiency. By which to compare all known functions, including EMD.

 
hrenfx:

Thanks for the article, enjoyed reading it.

Thanks for your interest in the article.

I'm sorry, but I won't be able to go into the details of EMD . I wanted to create a software implementation of this method. Such an implementation was done, which was the basis for writing this article. No serious research has been done.

On EMD:

Does the EMD result depend strongly on the type of spline function used?

I have not tried to use other splines, for example of the fourth degree, so I do not have my own opinion about it. I think I've seen publications on this subject somewhere, but unfortunately I don't remember exactly where.

Please comment on the reason why EMD has chosen such a construction criterion:

At any point in the empirical mode, the mean of the envelopes defined by local maxima and local minima must be zero.

I can't comment here, it should be looked up in Huang's papers. These are his conditions.

 

The study of any decomposition method requires software implementation, so the basis is almost completely disclosed in the paper.

P.S. It is strange that you have never come across criteria for comparing different functions anywhere (on the Internet). This is a reason to think about your bicycle.

 

This is supposed to be an aternative and better method :Hilbert Vibration Decomposing  http://hitech.technion.ac.il/feldman/hvd.html  Maybe author or somebody with electronic engineering background write a new article.

 

 

First of all, thank you so much to the author for the article! It is interesting, clear and concise. The material is just enough to understand whether you need it or not and to decide whether to continue studying this topic further or not - a kind of introductory lecture. Besides, the author gave a base for initial testing. For this I would like to thank you very much.

Now, I will address those who need everything at once and at the same time, so that they don't need to move a finger. People - switch on your brains and fight your laziness. Some posts are simply unpleasant to read. The author tried hard, and you do not appreciate.

 
MetaQuotes Software Corp.:

New article Introduction to Empirical Modal Decomposition has been posted:

Author: Victor Likes

12345
New comment