[R] Defining time series objects

Paul Gilbert pgilbert at bank-banque-canada.ca
Wed Oct 31 16:28:01 CET 2001

>I am new to R, having used S-plus a number of years back.
>I would like to set up a time series object for forecasting, with data
>collected daily between 5th April 2001 and 16th September 2001.
>Any help would be most appreciated as I have been unable to find any
>suitable examples in the documentation.

If you are considering multivariate ARMA or state space models then you
might look at the dse bundle of packages on CRAN. You might also find
dse useful if you are looking at multi-period forecasts from univariate

There is a new version of dse on CRAN, but I have not widely advertised
it as I intend to soon clean up a couple of small glitches. In
particular, I mistakenly put the Users' Guide (which has lots of
forecasting examples) in the bundle subdirectory inst/doc rather than
dse1/inst/doc. Thus, it is distributed with the bundle but does not get
installed in an obvious place.

I have not used dse very much with daily data. Although it should not
make any difference, one is sometimes surprised. Please let me know if
you have any difficulty.

Paul Gilbert

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