Trading Strategies Based On Digital Filters - page 23
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
how do you guys want to break it up? cl
I'd like to focus on GBPUSD since that's where I've already started on
dee
FFL,
if you have an interest in catching up with us, you can easily do so by reading some past threads.
Dvarrin gave this thread a kickstart again on post #117. If you start from there you will get in on Simba, Clahn04 and Dvarrin's dialogue of what to do and why.
I also made a post on #191 that details what I did to build my indicators to reflect the latest market activity.
If you know how to Cut & Paste, and hit a button that says compile, then you pretty much can create your own updated indicators. With that said, you will need to read the previous posts to know what inputs to use and what to cut & paste. Overall it's not too difficult.
Why don't you start. Post pics if you get stuck and I'm sure we'll be able to walk you through. You might want to do what I did and write out why you did what you did and in what order so that someone can see if your thought process is on target.
have fun!DeeForex,
I am very much interested in catching up. Thank you for the referral to previous links. This will be a seemless "weekend" process.
Preciate the links to the post most important to getting up to speed. It seems they (the links) were out of sync w/ actual post to which you were referring. I don't say it to be an @hole. Just FEI (For Everyones Info) in case someone else comes along and gets confused.
Thanks for the warm welcome. As of late, most of TSD is some b.s. however there are still a few good threads out there because of a few good people remaining on them, case in point.
I am one to be very methodical and thorough w/ my reserach. To others I seem SLOW but I just like to understand the in's and out's to the point where I know what the heck I'm talking about vs. regergitating info. Reading the latest posts, it seems that everyone currently involved is taking it slow so as not to miss anything. Great minds think alike while simultaneously differently.
I have some ideas that seem pretty good. Looking forward to what the future holds.
Peace,
F.F.L.
Simba,
Could you post a simple synopsis of each mod and it's rules?
Thanks,
clAll use the custom SATL and RSTL I posted before..NO STANDARD SATL OR RSTL...No mod uses RSTL slope changes FROM POSITIVE TO NEGATIVE..both for entries and exits.
Mod1,catches retracements in a trend.. uses LOWRSTL and RSTL slope negative for sell..and closes by change of slope of RSTL.
Mod2 is the same as mod1 BUT uses,instead of RSTL a delayed SATL,delayed by 4 periods(check the code,I am talking by memory)that I forgot to attach and now attach here
Goertzel
Hi!
Here is a usefull (I hope) tool for realtime cycles computing I have just coded.
It's based on goertzel algorithm.
There are three values displayed:
index 0: Amplitude (not the squared one) in Red
index 1: phase (radians) in orange
index 2: Period value / 10^4 when it's a peak. 0 instead. (not drawn)
It computes fast (on my 1 year old computer) and there are three parameters:
MinPer (2 minimum) sets the minimum period analysed
MaxPer sets the maximum period analysed
bar2update sets the number of bars between recalculation.
Take care, you need at least 3*MaxPer bars to compute the indicator, otherless you will have a message in the experts log and nothing in the indi window.
I know some will be angry to only find the ex4 but I prefer you give me advice and needs of modification, I do it and when everybody ok i post the mq4.
I prefer to keep my code clean of modifications and the algorithm is easy, just google goertzel and you'll find it.
jojo
Edit: Bug on Phase has been corrected. Now it's between 0 and 2*Pi (better...). I did not change the graph.Jojo,Thanks a lot,will test on Monday.Goertzel adaptive algorithm is usually better than Maximum Entrophy to detect cycles in noisy time series..like in financial markets,for example
..so,it will be a wonderful addition to our toolskit,specially if we can reestimate cycles in real time.
Could you please give a more detailed explanation of its potential uses?
From the papers at Meyers analytics,it looks like the possibilities are extraordinary,from creating adaptive composite cycles,self modifying in real time,to create adaptive indicators,like Stlm2,etc,not limited to a single cycle..
jojo, WOW!
This sounds very powerful, although a bit over my head. Time to google and read some more. Can't wait to read the dialogue that will hopefully come from it.
thanks!
Hi!
Here is a usefull (I hope) tool for realtime cycles computing I have just coded.
It's based on goertzel algorithm.
There are three values displayed:
index 0: Amplitude (not the squared one) in Red
index 1: phase (radians) in orange
index 2: Period value / 10^4 when it's a peak. 0 instead. (not drawn)
It computes fast (on my 1 year old computer) and there are three parameters:
MinPer (2 minimum) sets the minimum period analysed
MaxPer sets the maximum period analysed
bar2update sets the number of bars between recalculation.
Take care, you need at least 3*MaxPer bars to compute the indicator, otherless you will have a message in the experts log and nothing in the indi window.
I know some will be angry to only find the ex4 but I prefer you give me advice and needs of modification, I do it and when everybody ok i post the mq4.
I prefer to keep my code clean of modifications and the algorithm is easy, just google goertzel and you'll find it.
jojo
Edit: Bug on Phase has been corrected. Now it's between 0 and 2*Pi (better...). I did not change the graph.Jojo,
A welcomed surprise. I was just chatting w/ Simba about such a setup @ automated cycle computing for the digital indicators. I have no idea how to apply it but I'm willing to learn.
So the red line is similar to the line in the Spectrum Analysis setup on the DIG software?
How does one use it to determine the cycles? I don't see any labelling so that we may identify peaks. Perhaps I'm trying to apply it in a manner in which it wasn't meant. Upon reading your post it seems that this is just the beginning and that you plan to automate the process. Man, so much to learn and so little patience with which to do it. lol Feel like a kid in a friggin candy store!
I don't know whether to ask questions or make suggestions (both of which will hopefully be intelligent). I think I will read more so that I may get a better understanding. Math was never my strongest subject, but doesn't have to stay that way.
F.F.L.
P.S.
That Meyers Analytic stuff is indeed amazing. Consistent performance using cyclic-based trading. Very promising. Thanks for the heads up. Will read also over the weekend.
*Edit*
NOW I know how to see the peaks. Must have data window open. Guess that's what happens when you just try things instead of asking questions a whole heap of questions. Thanks for this JoJo. It's beginning to make sense as to how this could be used to automate the optimization process for digital filters. Will have to wait to see how the generation of the coefficients, etc will be developed
Seems that the Fin-Ware guys have released an updated version of the DFG software. Looks good. Contacted them to inquire the price. Supposedly, they're offering a demo version of both the DFG & Spectral Analyzer softwares. Will keep you guys posted.
Goertzel_Cycle indicator
Hi!
Here is a usefull (I hope) tool for realtime cycles computing I have just coded.
It's based on goertzel algorithm.
There are three values displayed:
index 0: Amplitude (not the squared one) in Red
index 1: phase (radians) in orange
index 2: Period value / 10^4 when it's a peak. 0 instead. (not drawn)
It computes fast (on my 1 year old computer) and there are three parameters:
MinPer (2 minimum) sets the minimum period analysed
MaxPer sets the maximum period analysed
bar2update sets the number of bars between recalculation.
Take care, you need at least 3*MaxPer bars to compute the indicator, otherless you will have a message in the experts log and nothing in the indi window.
I know some will be angry to only find the ex4 but I prefer you give me advice and needs of modification, I do it and when everybody ok i post the mq4.
I prefer to keep my code clean of modifications and the algorithm is easy, just google goertzel and you'll find it.
jojo
Edit: Bug on Phase has been corrected. Now it's between 0 and 2*Pi (better...). I did not change the graph.
EDIT2: find last version on this post (231 next page) https://www.mql5.com/en/forum/173071
Goertzel cycle V1
I you want to compare Goertzel_cycles indi with fin-ware analyser, you must keep in mind that fin-ware displays the squared amplitude (because it relates to cycle power) and I just display the amplitude.
But if I display the squared one, you can't see the phase because values are too little.
So find attached Goertzel_Cycle_V1 with an option to display Amp^2 or just amp. Default display is now Amp^2.
jojo
ps:I delete last ex4 (init version)