RcppHMM: Rcpp Hidden Markov Model
Collection of functions to evaluate sequences, decode hidden states and estimate parameters from a single or multiple sequences of a discrete time Hidden Markov Model. The observed values can be modeled by a multinomial distribution for categorical/labeled emissions, a mixture of Gaussians for continuous data and also a mixture of Poissons for discrete values. It includes functions for random initialization, simulation, backward or forward sequence evaluation, Viterbi or forward-backward decoding and parameter estimation using an Expectation-Maximization approach.
Documentation:
Downloads:
Reverse dependencies:
Linking:
Please use the canonical form
https://CRAN.R-project.org/package=RcppHMM
to link to this page.