Hidden Markov Model (HMM) R Package Development

Introduction

This project involves the development of an R package for Hidden Markov Models (HMMs) in longitudinal data analysis. The package, named regmhmm, incorporates features for flexible modeling, variable selection using regularization techniques, and efficient computation with the Expectation-Maximization (EM) algorithm.

Features

• Flexible Modeling: Build HMMs and mixed HMMs for longitudinal data analysis.
• Variable Selection: Perform variable selection within HMMs using regularization.
• Efficient Computation: Utilize the Expectation-Maximization (EM) algorithm with coordinate descent for optimal computational performance.

Installation

# Install the development version from GitHub:
devtools::install_github("HenryLeongStat/regmhmm")