- Search Forex Factory
- 32 Results (3 Threads , 29 Replies )
- MisterH replied Mar 1, 2011
how about this? No, it would be good to find an expert. But they probably won't share.. so .. the usual suspects -> refereed papers, A1 international journals on signal processing, ...
- MisterH replied Mar 1, 2011
wavelets — I absolutely agree. The same is true for Intrinsic Time-scale Decomposition. But then the question is: which signal processing techniques are suited for online signal processing that give a relatively smooth outcome? - DFT (not for ...
- MisterH replied Jan 17, 2011
Tradency — Hi Forexcube, I think that you may be right that they do get pips for every trade... But what I'm more interested in is to see some consistently profitable systems on their platform: it's true that there are >600 systems (and manual ...
- MisterH replied Dec 9, 2010
Just a note: If you want to quickly see if an indicator is repainting or not, you can make an empty EA (just take the standard template with nothing in it), and do a visual mode back test on every tick. You can then load any template or indicator ...
- MisterH replied Dec 7, 2010
Somebody please explain me the use of non-causal indicators. And no, smoothing the end-point with the JJMA doesn't work.
- MisterH replied Nov 25, 2010
Machine Learning to train a causal filter? — Has anyone ever tried a machine learning approach to find a model the gives +/- the same shape as the non-causal smoothers (Hodrick-Prescott, SSA, ...). The inputs of the model would be a very good ...
- MisterH replied Nov 11, 2010
DSP is useless — For me this thread has been an absolute waste of time. Nothing really beats a JMA to smooth a time series (I mean the Tradestation version from Jurik Research). I challenge anyone who reads this to show a better filter than ...
- MisterH replied Jan 3, 2010
and the author is ... — look what I just found: url
- MisterH replied Dec 11, 2009
Corona Cycle — Take a look at Ehlers' corona cycle period vs. his homodyne discriminator. There's quite a difference...
- MisterH replied Dec 10, 2009
On page 121 of his book, "Cybernetic Analysis for Stocks and Futures" Ehlers states: "The lag of measuring the dominant cycle period is about eight bars." In 2004 He called it the most responsive technique available. So if Ehlers gave up on ...
- MisterH replied Dec 10, 2009
Nonlinear & Nonstationary DSP — I think by now it's pretty clear that DFT, Goertzel, ... are pretty much useless on fx time series. The problem is chronomorphicity. It's a difficult word for saying that things change over time: volatility, ...
- MisterH replied Nov 11, 2009
nonlag filter example — MadCow , If this is something you are interested in, you can find it on another forum; I'll send it to you if you need it; I can't post it because it is not in the public domain.
- MisterH replied Nov 7, 2009
Forget about Bartels — Forget about Bartels. It is of no use in this context.
- MisterH replied Nov 6, 2009
AMI & time frame — Actually, H1 data have shown persistent trends (H = +/- 0.6), indicating that trend following strategies should outperform mean-reversion trading systems. I don't have that UBS study on AMI & time frames and I forgot its ...
- MisterH replied Nov 5, 2009
fractal dimension of different time frames — In theory yes, all time frames are supposed to be self-similar, but plot a couple of histograms and you well see that the lepto-kurtosis (heavy tails of the probability distribution) increases as ...
- MisterH replied Nov 5, 2009
some more .. — after looking some more I found these 2 on my laptop: "The Very Fast Fourier Transform ~ A Tool for Science in the 21st Century" and "Design and development of a high-speed winograd fast fourier transform processor board" url ...
- MisterH replied Nov 5, 2009
A lower complexity discrete fourier transform — Here is the complete article: "A lower complexity discrete fourier transform" It's from the British library, and they protect their content with a special reader.
- MisterH replied Nov 5, 2009
papers CB reffered to — Hi, pls find the 2 papers/articles CB referred to attached. If needed I can also upload the less critical papers. If anyone can get DCT in mq4 pls post. cheers
- MisterH replied Oct 15, 2009
Great work asm8086, how would you increase the smoothness of your filter to make it resemble the Hodrick-Prescott filter even further? Use a smoothing method like in this article or simply increase the length of the window it's calculated on?
- Length of the window to evaluate performance
I'm looking for ideas on how to decide what the appropriate size would be of the window to evaluate ...