Edge effect on the way to the GRAAL - page 3
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
The non-symmetric function, as I understand it, builds only on the previous values of the time series.
The edge effects are created by the last line.
Very nice indicator. But it has either distortions or lags again.
How to trade using it is a question.
Tomorrow is the holiday, I wish everyone Happy New Year of the Red Bull.
MAY OUR BALANCE SHEETS BE BULLISH !!!! :)
I will start working with Matlab from the first numbers, if anything turns up, I will post in this forum.
Good luck everyone and thanks.
p.s. For future posts: throw in ideas for reducing edge effects and smoothing functions.
Maybe together we'll find a grail :).
Once again, happy NEW YEAR!
It will work))) it's not going anywhere.
Dobeshi, 5 order of magnitude 3. From Matlab
It will work))) it's not going anywhere.
Dobeshi, 5 order of magnitude 3. From Matlab.
On outcomes further than 10 from the current point everything is beautiful. Question how the wavelet behaves on the last outcome.
Just got online after the holidays. Still confusing the keys :)
I tried Dobesh 7 order and level 5. True for M1.
VladEvgeniy, I understand that you managed to compile and connect the library from Matlab to MT.
Would you share the DLL? I haven't got to a disk shop yet and it's hard to download 4.5 Gbytes.
I will try to connect the library to my Expert Advisor and test it with a new smoothing algorithm.
Without matlab, the dll will not work. You need the matlab installed, or the libraries separately (they are in the matlab folder). The dll itself won't do anything. In addition, you can first look at the matlab itself and assess whether it is worth the hassle or not.
Eh... I'll go get the disc.
The case is unsuccessful and is a consequence of the fundamental impossibility of looking into the future. Therefore, the marginal effect is unavoidable in principle irrespective of the smoothing method, but it can be reduced to a theoretical minimum, which, however, does not give a noticeable trading advantage.
I agree. For some reason most people try to trade through knowingly losing systems based on indicators.
Looking for patterns in comparing price behavior and some function of this price (indicator) coming up with filters as exclusion rules - does not seem to be the right way to a profitable system.
If we look for patterns, we should look for them in the price itself.
I agree. This is why most of them try to trade using knowingly losing systems based on indicators.
Looking for patterns in comparing the behavior of price and some function of that price (indicator) coming up with filters as exclusion rules - is not the right way to a profitable system.
If we are searching for patterns, then in the price itself.
Of course, the original source is the price. But an indicator is a transformed price, from which unnecessary information has been discarded.
The price regularities are also preserved in the indicators. Correlation analysis can confirm it.
The proof of the profitability of indicator systems can be found in the long-term worldwide experience of trading on indicators.
You will not reject it, right? At least, the normalized price indicator (by expected payoff and dispersion) already contains
It has a lot of interesting things and shows what is not visible on the price chart.
I agree about filters. The system must be one, without any patches.
I used to think so too, until very recently. But recently I've been looking the other way. To make a long story short, you can do the trick here. The filter itself can, under certain conditions, be formally swapped with the main indikator, thus calling the indikator a filter. This may drastically affect the interpretation of the system itself. But this is more in the realm of mind games.
Installed Matlab 7.01. Powerful stuff.
Found the wavelets.
But how do I load my signal into the system?
Is there a converter for example from a text file to a MAT file?