R package for statistical inference using panel POMPs (Partially Observed Markov Processes)

View the Project on GitHub cbreto/panelPomp

Package description

Tools for working with longitudinal or panel POMPs, i.e., with Partially Observed Markov Processes (AKA state-space models, stochastic dynamical systems) involving multiple, independent units (or individuals) for each of which (potentially multivariate) time series data is available. The basic idea driving 'panelPomp' is to apply to a collection of units (or individuals) some of the 'pomp' package facilities for implementing POMP models, simulating them, and fitting them to time series data. Regarding fitting, only the iterated filtering (mif2) algorithm has currently been extended to the panel setting. (Alpha release version.)

Authors and Contributors

Authors: Carles Breto (@cbreto), Edward L. Ionides (@ionides) and Aaron A. King (@kingaa).