[R] AR models

Prof Brian D Ripley ripley at stats.ox.ac.uk
Sat Sep 4 09:23:45 CEST 1999

On Sat, 4 Sep 1999, Michael HART wrote:

> Thanks for all who have helped me with my original queries, below.
> I have now managed to import some data and use the ar modeling.  Now I have
> some more specific questions about this package if anyone knows the answers
> to I would appreciate some help in clearing up a few issues.
> I used the routine ar.yw (amongst others) in the library ts.  This routine
> automagically chose the ar model order for me.  I was wondering what method
> it uses to do this?

You could try reading the help page:

     aic: Logical flag.  If `TRUE' then the Akaike Informa-
          tion Criterion is used to choose the order of the
          autoregressive model. If `FALSE', the model of
          order `order.max' is fitted.
     Order selection is done by AIC if `aic' is true. This

And AIC is described in the first, second and fourth reference, at least.

> The package allows the user to enter a maximum model order from which it
> chooses the model order it prefers (below this value).  However I would like
> to force it to use a particular ar model order of my choice as I know how
> many theoretical processes are in the data (which is more that the order
> being chosen).

ar(x, aic=F, order=3), for example.

> Can someone with experience with this package or with ar modelling tell me
> how to make it force a particular model order.

Advice: do read the help pages on the functions you use, especially before
asking here.

Library ts is close to but not the same as the functions in S-PLUS. These
and the backround are described in Venables & Ripley (1999) `Modern Applied
Statistics with S-PLUS' and how to do its examples in R is in the R version
of the script in the VR61* package on CRAN.

> Also the libraries ts and tseries each use the same name for their base data
> types but they are incompatible with each other.

We know they overlap, but I was not aware of incompatibility of the base
type (ts) which is defined in base R, not in any package. Package tseries
will be updated shortly: its author contributed some of the functionality
in package ts.

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 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

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 at stat.math.ethz.ch

More information about the R-help mailing list