FlexMix: A general framework for finite mixture models and latent class regression in R

By Friedrich Leisch. 2004

780 citations recorded by the source corpus.

FlexMix implements a general framework for fitting discrete mixtures of regression models in the R statistical computing environment: three variants of the EM algorithm can be used for parameter estimation, regressors and responses may be multivariate with arbitrary dimension, data may be grouped, e.g., to account for multiple observations per individual, the usual formula interface of the S language is used for convenient model specification, and a modular concept of driver functions allows to interface many different types of regression models. Existing drivers implement mixtures of standard linear models, generalized linear models and model-based clustering. FlexMix provides the E-step and all data handling, while the M-step can be supplied by the user to easily define new models.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/628379f1b7de8ef2d95b136e89859435b4293887

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

Model-Based Clustering (2016, background): “Indeed, as distinct from finite mixtures of regressions (e.g., Leisch 2004; Früwirth-Schnatter 2006), which are examples of mixture models with fixed covariates, CWMs allow for assignment dependence, i.e., the covariate in each component can also be distinct.”
Information Technology Competencies, Organizational Agility, and Firm Performance: Enabling and Facilitating Roles (2013, background): “…not identiﬁed, (1) multiple regimes may have similar estimates of the parameters of interest, which confounds the interpretation of results, and/or (2) there may be at least one regime to which extremely few ﬁrms (close to 0) can be assigned on the basis of posterior probabilities (Leisch 2004).”
Team Faultline Measures (2013, methodology): “Our implementation of the LCCA approach builds on the FlexMix package (Grün & Leisch, 2007, 2008; Leisch, 2004), which provides LCA in R.”
Customer segmentation using unobserved heterogeneity in the perceived value - loyalty-intentions link (2014, methodology): “The finite mixture model for all K segments k is then (Leisch, 2004)”
poLCA: An R Package for Polytomous Variable Latent Class Analysis (2011, methodology): “The regmix function in package fpc (Hennig 2010) will estimate this other type of LCR model, as will the flexmix function in package flexmix (Leisch 2004; Grün and Leisch 2008).”
Regression Models for Count Data in R (2008, methodology): “Some important model classes include ﬁnite mixture models—implemented in R in package ﬂexmix (Leisch 2004)—and generalized estimating equations (GEE)—provided in R by package geepack (Halekoh, Højsgaard, and Yan 2006)—and mixed-eﬀects models—available in R in packages lme4 and nlme (see Pinheiro…”
On‐line expectation–maximization algorithm for latent data models (2007, methodology): “6 .”
Estimation of extended mixed models using latent classes and latent processes: the R package lcmm (2015, methodology): “multivariate longitudinal data based on MCMC estimation; package FlexMix (Leisch 2004; Grün and Leisch 2008) also proposes the estimation of classes of mixture models including the latent class linear mixed model with function FLXMRlmer; and mixtools (Benaglia, Chauveau, Hunter, and Young 2009) incl”
ℓ1-penalization for mixture regression models (2010, methodology): “For computing the unpenalized MLE, we used the R-package FlexMix (Leisch, 2004; Grün and Leisch, 2007, 2008); Simulation 2 explores sparsity; Simulation 3 compares cross-validation and BIC; and Simulation 4 compares the different penalty functions with the parameters γ = 0, 1/2, 1.”
Visualizing Count Data Regressions Using Rootograms (2016, methodology): “3-13 (Leisch 2004; Grün and Leisch 2008), and were identical on various platforms including PCs running Debian GNU/Linux (with a 3.”
Modelling beyond regression functions: an application of multimodal regression to speed–flow data (2006, methodology): “…of generalized linear models (Wedel and Kamakura, 1995) (a simple tool to fit models of this type is the R package flexmix which was developed by Leisch (2004)); (b) ‘multiple-model regression estimation’ by Cherkassy and Ma (2005) (this fits parametric regression models consisting of several…”
mixtools: An R Package for Analyzing Mixture Models (2009, methodology): “This is the main feature that distinguishes mixtools from other mixture-related R packages, also available from the Comprehensive R Archive Network at http://CRAN.R-project.org/ , such as mclust (Fraley and Raftery 2009) and ﬂexmix (Leisch 2004; Gr¨un and Leisch 2008).”
Slow‐growing species cope best with drought: evidence from long‐term measurements in a tropical semi‐deciduous moist forest of Central Africa (2013, methodology): “codes based on the flexmix (Leisch 2004) and glmnet (Friedman,”