- Search Metals Mine
- 23 Results
- BarHopper replied Mar 5, 2009
Check your frequency vector — Hi Ngawang, I quickly glanced at your code, so take this with a grain of salt, but I think I see the problem. You're defining your frequency vector as: f1 = ((0:2:N)') ./ N ./ 2; while the FFT defines the ...
- BarHopper replied Feb 19, 2009
Matlab integration — Thanks BlackMage, Have you used any of these Matlab integration tools yourself? I've personally used MatlabTrader's interface to OpenTick. It mostly worked, but OpenTick was very badly supported and I was wasting a lot of ...
- BarHopper replied Feb 19, 2009
Matlab and brokers — Hi BlackMage, Matlab is by far my favorite development/programming environment... but I'm not aware of the ways it can be interfaced to brokers. I'd love to learn about the options. Would you mind sharing what you know? ...
- BarHopper replied Feb 19, 2009
Nice curve — Hi CB, What strikes me about your equity curve is how linear it is... very nice. IMHO, consistency is even more important than average profit, and your system has both. With a system this consistent you can do whatever you want. ...
- BarHopper replied Feb 17, 2009
Anyone care about missing bars? — Yes mikkom, I assumed the bet size was static based on 5% of the initial deposit. Would you do that in practice? Maybe, if you made a practice of frequently withdrawing or paying out your profits, as an ...
- BarHopper replied Feb 17, 2009
Money management — Hi Julek, Yes, you are correct, those simulations were done with no money management whatsoever... I've edited my post to clarify this. By "underwater," I'm not saying that you'll lose all your capital, just that you'll lose ...
- BarHopper replied Feb 16, 2009
Probability of losing all your winnings... — Thanks for the link Sixer... Those were good presentations. However I did find a glaring error (both conceptual and mechanical) in Ehlers' slide on Bertrand's Ballot Theorem. Did anyone else see ...
- BarHopper replied Feb 16, 2009
Missing bars... — Hi all, This is a question that comes up all the time with intraday timeframes... If the bar period is short and/or the trading is thin, there will be missing bars from time to time. I wonder what your attitudes are in ...
- BarHopper replied Feb 15, 2009
Some like coffee... — With all due respect, from the time Codebreaker first started this thread, the whole point has been the application of signal processing to pro-trend trading systems. Let's try to keep the focus and let other threads ...
- BarHopper replied Feb 15, 2009
Thanks for the code... — Thanks CB and MisterH for the Matlab code... (Thanks to you too anx, but I may be the only one here without MT4, so I can't use it.) It never ceases to amaze me how much advanced technology is just out there, waiting ...
- BarHopper replied Feb 14, 2009
amost equal — Hi CB, The only difference is that I have s1 in the imagPart where your 2nd mathcad diagram had s2. It was probably just a typo. Otherwise they are the same. BTW, changing gears a bit... you seem to like the SSA algorithm... I've ...
- BarHopper replied Feb 14, 2009
Goertzel revisions — Hi CB, I just tried your revised equation for the Goertzel algorithm, and I get a significant disparity from the FFT. Perhaps we're using different definitions of loop counters or something? If you like, you can look at my ...
- BarHopper replied Feb 13, 2009
Goertzel == FFT — Well, I think I've proven it to myself, the FFT and Goertzel algorithm give results which are identical within the precision of my machine. This applies to magnitude and phase, on a number of test signals containing ...
- BarHopper replied Feb 13, 2009
Musings on information and chaos — Well I'm not sure how deeply I've been able to go in information theory, but I've thought about it a bit. I like the various entropy measurements because they summarize non-linear and non-obvious ...
- BarHopper replied Feb 13, 2009
Organizing papers — BlackMage: I ran into this "too many papers" problem a few years ago when I was working for a company with several overlapping R&D projects going on at once. We never found a good multi-person solution. My personal ...
- BarHopper replied Feb 12, 2009
Possibly dumb question — To whomever has implemented and understood the benefits of the Goertzel algorithm: I'm a bit confused... My understanding is that the Goertzel algorithm is one implementation of the DFT, favored in embedded ...
- BarHopper replied Feb 12, 2009
FNN vs. MI? — Thanks CB for that... but to quickly follow up, do you have any thoughts on the pros / cons of using FNN instead of MI to determine the timeframe? Thanks...
- BarHopper replied Feb 12, 2009
FNN, timeframes, and MI — Thanks CB, I think I understand... as long as your D is big enough to "unfold" the dynamics, you can hold it fixed, and construct a plot of FNN% vs. timeframe, and pick the minimum. This is instead of plotting the ...
- BarHopper replied Feb 12, 2009
Timeframes and embedding — Hi all, I would like to second the question by chamois (post 238). CB, in my attempt to understand your post 230, I've done a little reading, and have come up with a few questions: My first question is to clarify ...
- BarHopper replied Feb 11, 2009
With all due respect — First some context: I've been following this thread as a background task for several months now... I've only posted a few times, while doing my own research (which only partly overlaps the content in this thread) ...