Learn R Programming

hmmhdd (version 1.0)

Hidden Markov Models for High Dimensional Data

Description

Some algorithms for the study of Hidden Markov Models for two different types of data. For the study of univariate and multivariate data in a finite framework, we provide some methods based on the definition of a Gaussian copula function to define the dependence between data (for further details, see Martino A., Guatteri, G. and Paganoni A. M. (2018) ). For the study of functional data, we define an objective function based on distances between random curves to define the emission functions of the HMM (for further details, see Martino A., Guatteri, G. and Paganoni A. M. (2019) ).

Copy Link

Version

Install

install.packages('hmmhdd')

Monthly Downloads

9

Version

1.0

License

GPL-3

Maintainer

Andrea Martino

Last Published

September 4th, 2019

Functions in hmmhdd (1.0)

set_fhmm

S3 Class for Hidden Markov Models with functional response.
set_mhmm

S3 Class for Hidden Markov Models with multivariate response.
simulatedFD

Simulated multivariate functional dataset with underlying Markov Model
bic

BIC Bayesian Information Criterion for the HMM
summary.fhmm

Summarizing functional Hidden Markov Models
viterbi

Viterbi Function
summary.mhmm

Summarizing multivariate Hidden Markov Models
rmdistr

Random generator Function for multivariate data
copulahmmdata

Simulated copula dataset with underlying Markov Model
forwardbackward

Forward-Backward Function
aic

AIC Aikake Information Criterion for the HMM
plot.fhmm

Plotting functional Hidden Markov Models
fitBM_mhmm

Baum-Welch Function for multivariate data
dmdistr

Copula density Function for multivariate data
copuladata

Simulated copula dataset
fitBM_fhmm

Baum-Welch Function for functional data