Is there any good on the NMA (aka moving average 3.0)...?

New comment
 

Hi fellows,

In this thread I would like to analyze if this new type of moving average may/may not to bring any good...

I'm showing here my first experiment with it; I did used like base, the Cuttler's RSI and I did replaced the SMA with the NMA.

The result was a kind of "extremely nervous" oscillator, so I did applied to the final signal another stage of smoothing, and attached you'll find the result. Considering that actually is a sort of "double smoothing" the RSI is still very fast, seems almost a sort of "gain in amplitude" of the signal without significant lag.

Your comments will be really appreciated and anybody can bring on the table his/her own modification/adds and, more important, new experiment!

In the picture you'll see the new RSI on NMA (Blue) superimposed to the regular (Red), both 14 periods; the other settings for the NMA are in the code.

P.S. this indi doesn't need the NMAv1.2 it's a standalone; Mladen or MrTools could you please add your "magic touch" and calculate the NMA without using buffers? (I still have a lot to learn about mql... )

the original NMA indicator is on this post https://www.mql5.com/en/forum/179807/page78

Best Regards

Files:
rsi_on_nma_v1.0.mq4  4 kb
rsionnma.gif  43 kb
 

...

brax64

Here you go all done in arrays, no buffers at all.

___________________________________________

There are slight differences in values. The difference comes from the way how metatrader treats integer value math operations. If you replace this line in the original :
double Lambda = NMA_Length / NMA_S_Length;[/PHP]
with this one :
[PHP] double Lambda = 1.0*NMA_Length / NMA_S_Length;

(so, only multiplication with 1.0 added, in order to force metatrader to use floating point calculation instead of integer calculation (since both NMA_Length and NMA_S_Length are integers) you are going to get exactly the same values of the two indicators. Which also means that the same line in nma itself should be altered in order to make it calculate in maximal precision. Quite frankly I am getting sick of this kind of math metatrader is doing and they did not provide us with any type cast modes in mql4 so we always have to look 10 times for things like this

regards

brax64:
Hi fellows,

In this thread I would like to analyze if this new type of moving average may/may not to bring any good...

I'm showing here my first experiment with it; I did used like base, the Cuttler's RSI and I did replaced the SMA with the NMA.

The result was a kind of "extremely nervous" oscillator, so I did applied to the final signal another stage of smoothing, and attached you'll find the result. Considering that actually is a sort of "double smoothing" the RSI is still very fast, seems almost a sort of "gain in amplitude" of the signal without significant lag.

Your comments will be really appreciated and anybody can bring on the table his/her own modification/adds and, more important, new experiment!

In the picture you'll see the new RSI on NMA (Blue) superimposed to the regular (Red), both 14 periods; the other settings for the NMA are in the code.

P.S. this indi doesn't need the NMAv1.2 it's a standalone; Mladen or MrTools could you please add your "magic touch" and calculate the NMA without using buffers? (I still have a lot to learn about mql... )

the original NMA indicator is on this post https://www.mql5.com/en/forum/179807/page78

Best Regards
Files:
rsi_on_nma_v1.01.mq4  6 kb
 

Hi Mladen,

thank you very much! Really appreciated your help and mentoring!!!!

Regards

mladen:
brax64

Here you go all done in arrays, no buffers at all.

___________________________________________

....

.....

......

regards
 

That error went unnoticed by me either in the nma, I only noticed it when I was comparing the rsi you made with the newer one and was surprised to see some differences. So : xspuadflj!!!! the integer disivion in mql 4

brax64:
Hi Mladen,

thank you very much! Really appreciated your help and mentoring!!!!

Regards
 

Nma macd?

Hey Brax this rsi looks pretty nice does this mean this type of moving average can also be applied to be it's own moving average convergence divergence oscillator?

 

Alright,alright!

wow tha'ts pretty cool just like that the nma in macd form!

 

Hi Maxwell10,

well, I suppose that it can be used like every other averages; so now we have also the MACD of NMA It seems faster (with "classic" parameter 12,26,9) respect to the regular MACD. But maybe better to try some comparation playing with the period lenghts...

Best Regards

UPDATE: Mladen kindly advised me that was a small error in the code; I did correct now the MACD. If somebody did already download the indicator please downland it again and accept my apologies about that.

brax64

Maxwell10:
Hey Brax this rsi looks pretty nice does this mean this type of moving average can also be applied to be it's own moving average convergence divergence oscillator?
Files:
macd_fo_nma.gif  47 kb
macd_of_nma_v1.0.mq4  9 kb
 

RSI On NMA Dynamic Zones?

Hey Brax I know this may sound a little premature being that the NMA was just introduced to us but we might as well move it along to see if this thing really has legs and a heart! Is it possible to turn this into anything similiar like the dynamic zone pa IFT of rsi smoothed alerts_line_arrows mq4? perhaps the option can be added or due to way indy is calculated a whole new dynamic zone pa IFT Rsi on NMA alerts_line_arrows mq4 can be made,and remember Brax I'm not quite sure if I'm jumping the gun but I have been doing alot of research and I'm searching for the perfect rsi type which is why I' also asked mladen to implement T3 into dynamic zones if possible,but when I finish experimenting I will bring along to show and tell, something,I think some might find useful, thanks Max

 

Progress

This MACD of NMA shows great potential!

Hope this thread moves along!!

I just want to express some encouragment since I cannot move it along being a programming flop!

Cheers,

DK

 
brax64:
Hi Maxwell10,

well, I suppose that it can be used like every other averages; so now we have also the MACD of NMA It seems faster (with "classic" parameter 12,26,9) respect to the regular MACD. But maybe better to try some comparation playing with the period lenghts...

Best Regards

UPDATE: Mladen kindly advised me that was a small error in the code; I did correct now the MACD. If somebody did already download the indicator please downland it again and accept my apologies about that.

brax64

Upgraded version of this macd of nma : macd_of_nma_v1.01.mq4

Files:
macd_of_nma_v1.01.mq4  9 kb
macd_of_nma.gif  78 kb
 
mladen:
Upgraded version of this macd of nma : macd_of_nma_v1.01.mq4

More upgrade done (to make it more user friendly) : macd_of_nma_v1.02.mq4

Files:
macd_of_nma_v1.02.mq4  9 kb
New comment