R help - May 2005 - ARIMA estimation

Good morning,

(sorry for my english)
i have some problems to put off by extimation ARIMA coefficients 
the ones not significatives.
 Exist a method to extimate only that significatives?

i use the command: arima().

thanks to all
Stefano
	[[alternative HTML version deleted]]

On Thu, 19 May 2005, Stefano Colucci wrote:
> Good morning,
>
> (sorry for my english)
> i have some problems to put off by extimation ARIMA coefficients
> the ones not significatives.
> Exist a method to extimate only that significatives?
>
> i use the command: arima().
See the argument 'fixed', which allows you to choose which coefficients
to
estimate.

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

Le 19.05.2005 09:23, Stefano Colucci a ??crit :
>Good morning,
>
>(sorry for my english)
>i have some problems to put off by extimation ARIMA coefficients 
>the ones not significatives.
> Exist a method to extimate only that significatives?
>
>i use the command: arima().
>
>thanks to all
>Stefano
>  
>
You can fix coefs to what you want by using the parameter fixed in the 
arima call.

arima(x,fixed=c(NA,0,NA),order=c(3,0,0))

will fit an arima(p=3,i=0,q=0) with \phi_2 fixed to zero.

Maybe you'll be interested in graph 29 in the graph gallery : 
http://addictedtor.free.fr/graphiques/
See
http://addictedtor.free.fr/graphiques/RGraphGallery.php?graph=29

Romain

-- 
~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~
~~~~~~      Romain FRANCOIS - http://addictedtor.free.fr         ~~~~~~
~~~~        Etudiant  ISUP - CS3 - Industrie et Services           ~~~~
~~                http://www.isup.cicrp.jussieu.fr/                  ~~
~~~~           Stagiaire INRIA Futurs - Equipe SELECT              ~~~~
~~~~~~   http://www.inria.fr/recherche/equipes/select.fr.html    ~~~~~~
~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~