pomp package provides facilities for inference on time series data using partially-observed Markov processes (AKA state-space models or nonlinear stochastic dynamical systems).
The user encodes a model as a pomp object by providing functions specifying some or all of the model's process and measurement components.
The package's algorithms for fitting models to data, simulating, etc. then call these functions.
At the moment, algorithms are provided for
particle filtering (AKA sequential Monte Carlo or sequential importance sampling, see pfilter),
the likelihood maximization by iterated filtering (MIF) method of Ionides, Breto, and King (PNAS, 103:18438-18443, 2006, see mif),
and the nonlinear forecasting algorithm of Kendall, Ellner, et al. (Ecol. Monog. 75:259-276, 2005, see nlf).
Future support for a variety of other algorithms is envisioned.
A working group of the National Center for Ecological Analysis and Synthesis (NCEAS), "Inference for Mechanistic Models", is currently implementing and testing additional methods for this package.The package is provided under the GNU Public License (GPL). Contributions are welcome, as are comments, suggestions for improvements, and bug reports.
pomp, is provided to encode a partially-observed Markov process together with a uni- or multi-variate data set.vignette("intro_to_pomp") or look at the HTML documentation to view the vignette.pomp, simulate-pomp, mif, nlf