hmm-make {stochmod}R Documentation

Hidden Markov Models

Description

Generate an instance of HMM parameters.

Usage

HMM.make( K, p, mu=rep(0, p), sigma=diag(p) )

Arguments

K Number of components
p Dimensionality
mu General center of the data being modeled
sigma General orientation of the data being modeled

Details

Component centers are generated from a multivariate Normal distribution with the provided mean mu and covariance matrix sigma. Component covariance matrices are all set to the provided sigma. Prior distribution is uniform across all states. Rows of the transition matrix are random multinomial distributions.

Value

A Hidden Markov Model defined by:

mu [K x p] matrix of component means
sigma [K x p x p] array of component covariance matrices
pi [K x 1] vector of mixture coefficients
A [K x K] state transition matrix with element (i,j) referring to transition from state i to state j

Author(s)

Artem Sokolov Artem.Sokolov@gmail.com


[Package stochmod version 1.2 Index]