Thursday, June 24, 2010

New paper on analysis of stable isotope data

Ward, E.J., Semmens, B.X., and D.E. Schindler. 2010. Including source uncertainty and prior information in the analysis of stable isotope mixing models. In press, Environmental Science & Technology. doi:10.1021/es100053v

Tuesday, June 22, 2010

MARSS package uploaded to CRAN

The official release of 1.0 has been uploaded to CRAN (Here's the link to the CRAN page). This is an R package to fit unconstrained and constrained multivariate autoregressive state-space models via maximum-likelihood (EM algorithm). It was developed by Eli Holmes, Eric Ward and Kellie Wills. It is fully documented with a user guide/manual: Analysis of multivariate time-series using the MARSS package, by E.E. Holmes and E.J. Ward. MARSS has bootstrap AICb and CIs and simulation features. MARSS 1.0 doesn't allow you to estimate the B matrix. MARSS 2.0 will allow that, and MARSS 3.0 will provide Bayesian fitting. If you are looking for the derivation of update equations for the Kalman-EM algorithm, take a look at Derivation of the EM algorithm for constrained and unconstrained multivariate autoregressive state-space (MARSS) models.