Global decoding for autoregressive HMM using Viterbi — viterbi

Global decoding for an autoregressive HMM with or without within-state autoregression using the Viterbi algorithm.

Usage

viterbi_arp(x, Gamma, delta, dists, autocor = 0, params, N, p)

x: Data matrix the model was fitted to.
Gamma: Transition probability matrix (full matrix, not only off diagonal entries).
delta: Stationary distribution.
dists: Vector of distributions in R-jargon.
autocor: list of autoregression vectors, respecting order of dists.
params: List of optimized parameters, returned by fitting the HMM.
N: Number of states.
p: Vector of degree of autoregression for each distribution (0 = no autocorrelation).

Estimated states using Viterbi.