[R-SIG-Finance] Estimate parameters of a Kalman Filter

[John C Frain]

There are two tasks involved.  The first is to set up the loglikelihood and the second is the optimisation.  The first is often more time consuming than the second.  They are well documented in the documentation for the packages listed in the various task views. You may find that the package dlmodeler of interest.  This package is a front end on the packages dlm, KFAS and FKF.  If you wish to do your own programming you would gain from looking at the way that they set up the various system matrices.

Best regards
John

Sent from my iPad

On 15 Jun 2011, at 04:52, Noah Silverman <noahsilverman at ucla.edu> wrote:

> This isn't entirely an "R" question, but I thought the group would be the best place to ask.
> 
> 
> I want to consider a Kalman filter on some time series data.  One of the hard parts is estimating the 4-5 parameters for the filter.  I assume some form of EM would be good, but am not clear on how to best implement it.  
> 
> There is probably an R package that will do this automatically, but I'd like to LEARN how to do this manually as it will lead to me developing some more advanced filters.
> 
> Any suggestions?
> 
> --
> Noah Silverman
> UCLA Department of Statistics
> 8117 Math Sciences Building
> Los Angeles, CA 90095
> 
> _______________________________________________
> R-SIG-Finance at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions should go.
>