Comprehensive and compact tutorial on Petris’ DLM package in R; with an update about Helske’s KFAS


A blogger named Lalas produced on Quantitative Thoughts a very comprehensive and compact tutorial on the R package dlm by Petris. I use dlm a lot.

Unfortunately, Lalas does not give details on how the SVD is used. They do report that their tutorial is based in part on slides by Petris, and on slides by Zivot and Yollin. Petris himself acknowledges the SVD approach as originating with:

  • L. Wang, G. Libert, P. Manneback, “Kalman Filter Algorithm Based on Singular Value Decomposition,” Proceedings of the 31st Conference on Decision and Control, 1992, pp. 1224–1229.
  • Y. Zhang, R. Li, “Fixed-Interval Smoothing Algorithm Based on Singular Value Decomposition,” Proceedings of the 1996 IEEE International Conference on Control Applications, 1996, pp. 916–921.

Update, 1st August 2015

While reading a review by Tusell, my attention was drawn to the very recent (2015) KFAS package, developed and described by Helske, which I’m intending to try as a competitor to dlm. The detailed references are:

About hypergeometric

See http://www.linkedin.com/in/deepdevelopment/ and http://667-per-cm.net
This entry was posted in Bayes, Bayesian, dynamic linear models, dynamical systems, forecasting, Kalman filter, mathematics, maths, multivariate statistics, numerical software, open source scientific software, prediction, R, Rauch-Tung-Striebel, state-space models, statistics, stochastic algorithms, SVD, time series. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s