Adaptive lookback indicators - page 25
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
Ocean
Thanks Mladen ! Appreciate your effort and time on this. I will try it out.
Another thing which I experimented and trying to figure what will the decision on trade when price reach 50%(middle). So I use elliotoscillator1.02 as confirmation. It is not 100% accurate but it will help and also elliotoscillator yellow bar will 'shift' when new low/high is achieved close to or when it has not completed the full 1-5 cycle count. But it is a good guide. If snr members hv better idea, it would be interesting to hear this.
regards
KLI'll have a look with Ocean; perhaps its 'non-arbitrariness' will play well here...
Thanks Mladen ! Appreciate your effort and time on this. I will try it out.
Another thing which I experimented and trying to figure what will the decision on trade when price reach 50%(middle). So I use elliotoscillator1.02 as confirmation. It is not 100% accurate but it will help and also elliotoscillator yellow bar will 'shift' when new low/high is achieved close to or when it has not completed the full 1-5 cycle count. But it is a good guide. If snr members hv better idea, it would be interesting to hear this.
regards
KLHi,
Could you attach the elliottoscillator 1.02? I did a search for it to no avail,this way I can test it and comment.
In the meantime ,I suggest you try the dpjwma oscillator by Mladen with 10,7 settings instead of the usual 26,12.
Find it attached here.
Regards
S
Hi,
Could you attach the elliottoscillator 1.02? I did a search for it to no avail,this way I can test it and comment.
In the meantime ,I suggest you try the dpjwma oscillator by Mladen with 10,7 settings instead of the usual 26,12.
Find it attached here.
Regards
SHi Simba,
See attached. I got it from here as well.
Best Regards
KLelliot_oscillator_-_waves_1.02.mq4
Hi,
Yes ,you may be right and I see that Mladen is trying to solve this issue,in the meantime,may I suggest you apply the 40 DeviationsPeriod,1 SwingCount settings to EURUSD M15 chart,instead of an H1 one?This way we can all check the differences due to different brokers.
Thanks in advance and Regards.
SHi Simba,
This is the bands with your settings on Alpari data,started using it with dwpma and jma, so far very good.
kokleongch
Try out this version. One option added : SkipSunday. If it is set to true, it will skip Sunday data in calculations (linear regression and deviations) and you should see a straight lines through Sunday data (in the example picture, see how the linear regression value continues to be a straight line regardless of the change in data). If it is set to false, it work the same as before.
PS: making all calculations (adaptive look back + linear regression + deviations) skip Sunday in this indicator was a rather interesting experience, I must say (it caused me a couple of head scratching hours
) But, it really seems that the indicator depends a lot on broker data (as far as I see, mainly the problem is where brokers makes swings, and they are not as similar as it seems on a first glance)
regards
MladenHi Mladen,
Taking out the Sunday data seemed to affect the indicator as well. Pls see attached. Somehow the data is needed but time stamp should not recognize it as Sunday ? I don't know the detail
.
Thanks
KL
kokleongch
As far as I see from the pictures it is OK. The indicator is not skipping Sunday, but is ignoring it. As you cann see on your chart linear regression value lines is perfectly straight where it is Sudnay (as it should be). I hope that I understood the question right and that this ianswers its
PS: Sunday int the indicator is broker Sunday (not local Sunday)
Hi Mladen,
Taking out the Sunday data seemed to affect the indicator as well. Pls see attached. Somehow the data is needed but time stamp should not recognize it as Sunday ? I don't know the detail
.
Thanks
KLHi Simba, This is the bands with your settings on Alpari data,started using it with dwpma and jma, so far very good.
Hi MrTools,
I was using Mladen and yours divergence version too,and I was thinking exactly the same,fantastic when there is a divergence and a band touch at the same bar or near.
I am testing several trend filters on a higher TF(H4) ,any suggestion?Of course a good trend filter may be the direction of the steps in alb Mogalef,but I like to test with higher TFs.
Regards
S
Simba
Admit it, you enjoy doing these kind of things to me
It is mtfing the future (again ). Except that this one can be solved in a much simpler manner (after all it is a simple linear regression slope projection to the future). So here it is: multi time frame and alerts added. Alerts work properly in mtf mode too. It will alert on every breakout - meaning that it will not wait for opposite direction breakout but will alert whenever upper or lower band is broken
regards
MladenThat is why I said "the usual culprits"
.
Thanks a lot for your effort and dedication,your work truly increases the degrees of "freedom for the people"
...You know what I mean and that I mean well 
.
Regards
S
Mtf
Mladen,MrTools
As you can see by the pic,MTFing the indicator allows for faster visual evaluation of specific cases where we may have even higher probabilities of finding strong support or resistance.
Please check the pic...Lower M15 band is =/very near M30 CenterLine,with same settings(40 deviations,1 swingcount)...and that is where we found the rebound.
Mladen,since this indicator is "statistically based and biased",a confluence of timeframes MAY have a similar effect as the confluence of simple rules within a complex rule that we were privately discussing yesterday,it may not add "squared" probability(actually squared probability of rebound not happening) like... If 1 signal is 60% probability of rebound(so,40% failure probability)Then 2 signals have 16% failure probability(0.4*0.4) hence the rebound is 84% probable...But It adds to the base probability,so we know that we have more than x%(60% in the example) and less than y%(84% in the example) of the rebound happening...Of course,once we have enough samples of the double event happening,we may conclude that the squared probability of error is indeed the rule....Till then,it suffices to say that a double event improves the success stats.
Enjoy the weekend both of you...and everybody else...we deserve it
Regards
S