If anyone is interested in Beta testing a newly developed Moving Average I would like to hear from you, particularly if you use moving averages in your trading and/or have some insight into them. I think you will find it interesting.
Respond here or amen.hammer@gmail.com
Pics were from Sunday evening E-mini trade --- description of MAs is on the "14 Per" chart
I should add that this is a preliminary to commercialization so it is a closed, read only study and will need your MC id which is used according to MC recommended procedures.
I have also added some of Monday's action in the euro
Beta testers wanted for a new very fast and stable MA
Beta testers wanted for a new very fast and stable MA
- Attachments
-
- EUR 21.28.png
- (60.07 KiB) Downloaded 371 times
-
- 28 Per.png
- MA 28 per Comp
- (104.4 KiB) Downloaded 386 times
-
- 14 Per.png
- MA Comp
- (94.59 KiB) Downloaded 432 times
Last edited by whammer on 05 Dec 2011, edited 5 times in total.
Re: Beta testers wanted for a new very fast and stable MA
Hello WhammerIf anyone is interested in Beta testing a newly developed Moving Average I would like to hear from you, particularly if you use moving averages in your trading and/or have some insight into them. I think you will find it interesting.
Respond here or amen.hammer@gmail.com
Pics were from Sunday evening E-mini trade --- description of MAs is on the "14 Per" chart
I should add that this is a preliminary to commercialization so it is a closed, read only study and will need your MC id which is used according to MC recommended procedures.
I have also added some of Monday's action in the euro
Could you explain briefly on how yours MA could be more effective than the other proposed?. While I don't use MA for the moment, I do have some backtest that I would like to be process using MA.
Have a good day
Martin
Re: Beta testers wanted for a new very fast and stable MA
Hi Martin - Do you have another MA in mind or are you speaking generally?
If general I can say that I have compared the one I developed to essentially all of the MAs that are in general use. The goal is to make a MA as smooth as possible while making tradeoffs with lag/speed, hence there is some subjective balancing between conflicting ends. While everyone associates MAs with crossover strategies, their real importance is their use as price proxies for many other indicators. From what I have seen, the two best MAs out there are first, Mark Jurick's (the JMA) and second, the Hull MA. Jurick's is an outstanding piece of work that is fast, smooth, with low lag. Second would be the Hull MA - also outstanding. I consider the T3, exponential, weighted , their doubles and triples, their derivatives , the TEMAs, DEMAs , etc etc to be fairly useless for any trading purpose. I think speed is essential, but without smoothness (which is really noise filtering) it is not usable. I believe the MA I have developed is stronger in all those areas than the two I mentioned and which I give high marks and use.
It is with great caution that I say that, but I do believe that is true.
I hope this answers some questions and would be happy to answer anything further.
If general I can say that I have compared the one I developed to essentially all of the MAs that are in general use. The goal is to make a MA as smooth as possible while making tradeoffs with lag/speed, hence there is some subjective balancing between conflicting ends. While everyone associates MAs with crossover strategies, their real importance is their use as price proxies for many other indicators. From what I have seen, the two best MAs out there are first, Mark Jurick's (the JMA) and second, the Hull MA. Jurick's is an outstanding piece of work that is fast, smooth, with low lag. Second would be the Hull MA - also outstanding. I consider the T3, exponential, weighted , their doubles and triples, their derivatives , the TEMAs, DEMAs , etc etc to be fairly useless for any trading purpose. I think speed is essential, but without smoothness (which is really noise filtering) it is not usable. I believe the MA I have developed is stronger in all those areas than the two I mentioned and which I give high marks and use.
It is with great caution that I say that, but I do believe that is true.
I hope this answers some questions and would be happy to answer anything further.
- TJ
- Posts: 7743
- Joined: 29 Aug 2006
- Location: Global Citizen
- Has thanked: 1033 times
- Been thanked: 2223 times
Re: Beta testers wanted for a new very fast and stable MA
here is a good collection of moving averages for you to compare... (zero-lag stuff too)
http://www.tradersxchange.com/viewforum.php?f=27
http://www.tradersxchange.com/viewforum.php?f=27
Re: Beta testers wanted for a new very fast and stable MA
Thank you Tj - I am always looking for new ideas. The first one I looked at and had to laugh was CSX-zero lag and it stated
"CSX is a zero-lag moving average that is far superior than AMA, and almost as good as the JMA from Jurik."
Anyway, I looked at the code and the reason it was better than average was it had adopted some ideas from the Hull MA, but it is not zero lag - there are no zero lag averages. The terminology comes from the title of one of Ehlers' papers "Zero Lag (Well, Almost)". People forget that critical "(Well, almost)", since Ehlers knows well there is no such thing.
"CSX is a zero-lag moving average that is far superior than AMA, and almost as good as the JMA from Jurik."
Anyway, I looked at the code and the reason it was better than average was it had adopted some ideas from the Hull MA, but it is not zero lag - there are no zero lag averages. The terminology comes from the title of one of Ehlers' papers "Zero Lag (Well, Almost)". People forget that critical "(Well, almost)", since Ehlers knows well there is no such thing.