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

[Noah Silverman]

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