AW: Re[2]: JMA (was: Hull Moving average)

["Volker Knapp" <volker@xxxxxxxxxxxxxx>]

I do not really intent to participate here too much but wouldn't it make
sense to see how MJ is trading with his own indicator and how successful he
is and then we know how to judge it?

Vk
 

-----Ursprüngliche Nachricht-----
Von: Adrian Pitt [mailto:apitt@xxxxxxxxxxxxx] 
Gesendet: Donnerstag, 11. Januar 2007 14:51
An: 'Roscoe'
Cc: 'IVT'; 'Alex Matulich'; omega-list@xxxxxxxxxx
Betreff: RE: Re[2]: JMA (was: Hull Moving average)

Roscoe,

I'm a bit confused.  While the JMA may have some complex maths behind it,
aren't we ultimately looking for an average (in whatever way we choose to
use it) that shows the least lag and the least noise, and Mark Jurik has
basically shown this to be the case with his JMA.  This would especially be
the case in markets with large gaps, and especially applicable to say
intraday charts and handling the move form one day to the next.    So I'm
not sure what you mean when you say it gets too far from what is happening??
Surely it does this less than any other moving average?  Or are you
basically suggesting that all moving averages fail to perform in any of your
tests?  That is a very different issue indeed.  One of the problems is that
a moving average can be used in so many different ways, and even if you
conducted hundreds of tests, it does not mean they aren't useful, as your
view on the world may have missed a key angle of perception that someone
else has seen and built into their modelling using moving averages.  

Now I have to ask another question :)  What on earth do you mean by
indicators that look the worst? That could mean almost anything :) 

Cheers,
Adrian

-----Original Message-----
From: Roscoe [mailto:Roscoe@xxxxxxxxxxxxxx]
Sent: Thursday, 11 January 2007 8:47 PM
To: Adrian Pitt
Cc: 'IVT'; 'Alex Matulich'; omega-list@xxxxxxxxxx
Subject: Re[2]: JMA (was: Hull Moving average)


Hello Adrian,

Thursday, January 11, 2007, 5:45:31 PM, you wrote:

> do you have a view on why the JMA toolset fails to outperform other 
> variations of price smoothing?

Too complex IMHO which gets too far away from what is actually happening.

FWIW the *worst* looking indicators usually perform the best.

Sorry but I can't give you a more philosophical answer - all I can do is
test and look at the results. FWIW I was rooting *FOR* JMA because it seemed
soooo good on paper but in the end.....

-- 
Best regards,
 Roscoe                          mail to: Roscoe@xxxxxxxxxxxxxx
-- 
2:52 PM

-- 
2:52 PM