README.md

In LeonieVm/lmfa: Latent Markov Factor Analysis for Exploring Measurement Model Differences in Intensive Longitudinal Data

lmfa package

lmfa allows to explore within-person changes and between person differences in measurement models in (intensive) longitudinal data by means of three-step continuous-time latent Markov factor analysis (LMFA).

Installation

You can download the development version from GitHub as follow:

install.packages("devtools"); library(devtools)

devtools::install_github("leonievm/lmfa")

Usage

After successful installation, you can perform LMFA by means of the three-step estimation. The package consists of three main functions that are shown below. For details about the function arguments, see the function documentations, which can be opened with ?functionname.

1. The step 1 function estimates the state-specific factor analysis models by means of an expectation maximization algorithm (with or without model selection):

step1(data,
      indicators,
      n_state = NULL,
      n_fact = NULL, 
      modelselection = FALSE, 
      n_state_range = NULL, 
      n_fact_range = NULL,
      n_starts = 25,
      n_initial_ite = 10,
      n_m_step = 10,
      em_tolerance = 1e-8, 
      m_step_tolerance = 1e-3, 
      max_iterations = 1000,
      n_mclust = 5)

1. The step 2 function obtains the posterior state-membership probabilities and the modal state assignments and calculates the classification error:

step2(data, model)

1. The step 3 function estimates the transitions between the states (conditional on covariates) by means of a continuous-time latent Markov model:

step3(data,
      identifier,
      n_state,
      postprobs,
      timeintervals = NULL,
      initialCovariates = NULL,
      transitionCovariates = NULL,
      n_starts = 25,
      n_initial_ite = 10,
      method = "BFGS",
      max_iterations = 10000,
      tolerance = 1e-10,
      scaling = "proxi")

Contribution

If you have any suggestions or if you found any bugs, please feel free to contact me via email.

LeonieVm/lmfa documentation built on Dec. 5, 2023, 1:38 p.m.