[R] state-space models and kalman filter

G.Janacek G.Janacek at uea.ac.uk
Thu Nov 9 10:17:55 CET 2000

>On Wednesday 08 November 2000 22:01, Michael Roberts wrote:
>>  Does anyone know if one can easily estimate state-space models
>>  using ML and the kalman filter using R?  I would be especially
>>  interested in a relatively flexible function that would allow for
>>  estimation
>>  of hyperparameters, or could be made to do so.
It is not difficult to do. There used to be an EM version in Splus( 
which ran on R pretty well out of the box) due to Shumway at Statlib. 
If it is gone try his book ot the paper in J.Time Series.

If you don't like  the EM algorithm its pretty simple to use Kalman 
for the loglikelihood in time series estimation ( ARMA /structural 
models) . I have several home brew variants - they are not amazingly 
fast but they seem to work OK as an input to nlm( the R I use is 
~0.62).  If anyone is interested I'm happy to pass them on. The usual 

   Dean				               e-mail: G.Janacek at uea.ac.uk
   School of Mathematics			          tel: 44-(0)1603-592849
   UEA, Norwich NR4 7TJ,UK		          fax: 44-(0)1603-593868
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