My email does not seem to receive any attention on R-help, so I
forward it on R-devel if someone has already faced the "problem".
Thanks in advance
Christophe
Début du message réexpédié :
> De : Christophe Dutang
> Date : 13 novembre 2009 23:30:14 HNEC
> À : r-h...@r-project.org
> Objet : AR
As you are sure of the accuracy of your code, why don't you tell me
where is my mistake?
Le 15 nov. 2009 à 12:03, Prof Brian Ripley a écrit :
--
Brian D. Ripley, rip...@stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford,
On Sun, 15 Nov 2009, Christophe Dutang wrote:
As you are sure of the accuracy of your code, why don't you tell me where is
my mistake?
Because your supervisor is paid to do that.
Le 15 nov. 2009 à 12:03, Prof Brian Ripley a écrit :
--
Brian D. Ripley, rip...@stats.ox.ac.
Try this:
debug(stats:::ar.yw.default)
and then run the ar function to step through the ar code so you can
see the results line by line and understand what it is doing at a very
detailed level.
On Sun, Nov 15, 2009 at 10:06 AM, Christophe Dutang wrote:
> As you are sure of the accuracy of your
Thanks Gabor,
My mistake was I used the 'cor' function rather than the function
'acf' with "correlation" argument.
Christophe
Le 15 nov. 2009 à 18:13, Gabor Grothendieck a écrit :
Try this:
debug(stats:::ar.yw.default)
and then run the ar function to step through the ar code so you can
s
