HMM-class: Class "HMM" to represent parameters associated with a...

In HDclust: Clustering High Dimensional Data with Hidden Markov Model on Variable Blocks

Description

An S4 class to represent the model parameters associated with one variable block in the HMM-VB. For brevity, we call this part of HMM-VB, specific to a particular variable block, an "HMM" for the block. New instances of the class are created by hmmvbTrain.

Methods

• show signature(object = "HMM") : show parameters of the HMM object.

• getPrenumst signature(object = "HMM") : accessor for 'prenumst' slot.

• getHmmParam signature(object = "HMM") : accessor for parameters of the HMM object. This function outputs a list with means, covariance matrices, inverse covarince matrices and logarithms of the determinants of the covariance matrices for all states of the HMM.

Slots

dim

Dimensionality of the data in HMM.

numst

An integer vector specifying the number of HMM states.

prenumst

An integer vector specifying the number of states of previous variable block HMM.

a00

Probabilities of HMM states.

a

Transition probability matrix from states in the previous variable block to the states in the current one.

mean

A numerical matrix with state means. kth row corresponds to the kth state.

sigma

A list containing the covariance matrices of states.

sigmaInv

A list containing the inverse covariance matrices of states.

sigmaDetLog

A vector with log(|sigma|) for each state.

HDclust documentation built on May 2, 2019, 9:20 a.m.