R-beta: arima

Ross & (ihaka@stat.auckland.ac.nz)
Tue, 15 Apr 1997 23:01:01 +1200

Date: Tue, 15 Apr 1997 23:01:01 +1200
Message-Id: <199704151101.XAA05657@stat9.stat.auckland.ac.nz>
From: Ross & <ihaka@stat.auckland.ac.nz>
To: r-help@stat.math.ethz.ch
Subject: R-beta: arima
In-Reply-To: <m3bu7hdafj.fsf@biopc43.uio.no>

Fredrik Glockner writes:
 > Has anybody ported the `arima'-package to R?  I'm specially interested
 > in the `arima.mle'-thingy.

There is a very big state-space package which Paul Gilbert has done.

I have made a start on converting some of the Applied Statistics code,
but it is only in a partially complete state.  I'm not teaching time
series this year so I'm unlikely to make much progress.

On the S front ...  Has anyone else had problems with arima.mle.
I tried it on standard examples from Box and Jenkins last year and got
very different results from what B&J and SAS ETS give.  I'm pretty
suspicious about it.
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request@stat.math.ethz.ch