Tools for working with partially observed Markov process (POMP) models (also known as stochastic dynamical systems, hidden Markov models, and nonlinear, non-Gaussian, state-space models). The package provides facilities for implementing POMP models, simulating them, and fitting them to time series data by a variety of frequentist and Bayesian methods. It is also a versatile platform for implementation of inference methods for general POMP models.
|Author||Aaron A. King [aut, cre], Edward L. Ionides [aut], Carles Breto [aut], Stephen P. Ellner [ctb], Matthew J. Ferrari [ctb], Bruce E. Kendall [ctb], Michael Lavine [ctb], Dao Nguyen [ctb], Daniel C. Reuman [ctb], Helen Wearing [ctb], Simon N. Wood [ctb], Sebastian Funk [ctb], Steven G. Johnson [ctb]|
|Date of publication||2016-11-23 00:32:52|
|Maintainer||Aaron A. King <email@example.com>|
|License||GPL (>= 2)|
abc: Estimation by approximate Bayesian computation (ABC)
bake: Tools for reproducible computations.
basic_probes: Some useful probes for partially-observed Markov processes
blowflies: Model for Nicholson's blowflies.
bsmc: The Liu and West Bayesian particle filter
bsplines: B-spline bases
dacca: Model of cholera transmission for historic Bengal.
design: Design matrices for pomp calculations
eulermultinom: The Euler-multinomial distributions and Gamma white-noise...
example: Examples of the construction of POMP models
gompertz: Gompertz model with log-normal observations.
kalman: Ensemble Kalman filters
logmeanexp: The log-mean-exp trick
lowlevel: pomp low-level interface
measles: Historical childhood disease incidence data
mif: Maximum likelihood by iterated filtering
mif2: IF2: Maximum likelihood by iterated, perturbed Bayes maps
nlf: Parameter estimation my maximum simulated quasi-likelihood...
ou2: Two-dimensional discrete-time Ornstein-Uhlenbeck process
package: Inference for partially observed Markov processes
parmat: Create a matrix of parameters
pfilter: Particle filter
pmcmc: The particle Markov chain Metropolis-Hastings algorithm
pomp: Constructor of the basic pomp object
pomp_fun: Definition and methods of the "pomp.fun" class
pomp_methods: Functions for manipulating, displaying, and extracting...
probe: Probe a partially-observed Markov process by computing...
proposals: MCMC proposal distributions
ricker: Ricker model with Poisson observations.
rw2: Two-dimensional random-walk process
sannbox: Simulated annealing with box constraints.
simulate_pomp: Simulations of a partially-observed Markov process
sir: Compartmental epidemiological models
spect: Power spectrum computation and spectrum-matching for...
traj_match: Parameter estimation by fitting the trajectory of a model's...