posterior_probs: Posterior Probabilities for (Mixture) Hidden Markov Models

In seqHMM: Mixture Hidden Markov Models for Social Sequence Data and Other Multivariate, Multichannel Categorical Time Series

Posterior Probabilities for (Mixture) Hidden Markov Models

Description

Function posterior_probs computes the posterior probabilities of hidden states of a (mixture) hidden Markov model.

Usage

posterior_probs(model, log_space = FALSE)

Arguments

model	A (mixture) hidden Markov model of class hmm or mhmm.
log_space	Compute posterior probabilities in logarithmic scale. The default is FALSE.

Value

Posterior probabilities. In case of multiple observations, these are computed independentlsy for each sequence.

Examples

# Load a pre-defined MHMM
data("mhmm_biofam")

# Compute posterior probabilities
pb <- posterior_probs(mhmm_biofam)

# Locally most probable states for the first subject:
pb[, , 1]

seqHMM documentation built on July 9, 2023, 6:35 p.m.