Three-step estimation of latent Markov models with covariates
Computational Statistics & Data AnalysisPublished 24 October 2014
Francesco Bartolucci, Giorgio E. Montanari, Silvia Pandolfi
Citations34
SJR quartileQ1
SJR score0.89
SNIP1.38
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A three-step approach to estimate latent Markov (LM) models for longitudinal data with and without covariates is proposed and the properties of the proposed estimator are illustrated theoretically and by a simulation study in which this estimator is compared with the full likelihood estimator.
Abstract
A three-step approach is proposed to estimate latent Markov (LM) models for longitudinal data with and without covariates. The approach is based on a preliminary clustering of sample units on the basis of time-specific responses only, and is particularly useful when a large number of response variables are observed at each time occasion. In such a context, full maximum likelihood estimation, which is typically based on the Expectation–Maximization algorithm, may have some drawbacks, essentially due to the presence of many local maxima of the model likelihood. Moreover, this algorithm may be particularly slow to converge, and may become unstable with complex LM models. The properties of the proposed estimator are illustrated theoretically and by a simulation study in which this estimator is compared with the full likelihood estimator. How reliable standard errors for the three-step parameter estimates are obtained is also shown. The approach is applied to the analysis of a dataset about the health status of elderly people resident in certain Italian nursing homes.
Keywords
Computer ScienceMathematics
Journal of the Royal Statistical Society Series B (Statistical Methodology)Maximum Likelihood from Incomplete Data Via the <i>EM</i> Algorithm
49,657 Citations1977A. P. Dempster, N. M. Laird +1 more
EconometricaA Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity
26,139 Citations1980Halbert White
Journal of the American Statistical AssociationStatistical Analysis With Missing Data
17,489 Citations1989Maureen Lahiff, Roderick J. A. Little +1 more
Wiley series in probability and statisticsFinite Mixture Models
7,410 Citations2018Geoffrey J. McLachlan, David Peel +1 more
The aim of this article is to provide an up-to-date account of the theory and methodological developments underlying the applications of finite mixture models.
Political AnalysisLatent Class Modeling with Covariates: Two Improved Three-Step Approaches
2,259 Citations2010Jeroen K. Vermunt
The correction method of Bolck, Croon, and Hagenaars is extended by showing that it involves maximizing a weighted log-likelihood function for clustered data, which makes it possible to apply the method not only with categorical but also with continuous explanatory variables.
Cambridge University Press eBooksApplied Latent Class Analysis
1,946 Citations2002Jacques A. Hagenaars, McCutcheon, Allan L. 1950-
BiometrikaExploratory latent structure analysis using both identifiable and unidentifiable models
1,585 Citations1974Leo A. Goodman
Political AnalysisEstimating Latent Structure Models with Categorical Variables: One-Step Versus Three-Step Estimators
1,265 Citations2004Annabel Bolck, Marcel A. Croon +1 more
Journal of Statistical SoftwareMulti-State Models for Panel Data: The<b>msm</b>Package for<i>R</i>
1,143 Citations2011Christopher Jackson
The range of Markov models and their extensions which can be fitted to panel-observed data, and their implementation in the msm package for R are reviewed, intended to be straightforward to use, flexible and comprehensively documented.
Hidden Markov Models for Time Series: An Introduction Using R
714 Citations2009Walter Zucchini, Iain L. MacDonald
Sociological MethodologyEstimating the Association between Latent Class Membership and External Variables Using Bias-adjusted Three-step Approaches
610 Citations2013Zsuzsa Bakk, Fetene Tekle +1 more
The results showed that also when the latent class variable is used as a predictor of external variables, the uncorrected three-step approach leads to severely biased estimates, and the proposed correction methods perform well under conditions encountered in practice.
The Annals of StatisticsIdentifiability of parameters in latent structure models with many observed variables
452 Citations2009Elizabeth S. Allman, Catherine Matias +1 more
A general approach for establishing identifiability utilizing algebraic arguments is demonstrated, which sheds light on the properties of finite mixtures of Bernoulli products, which have been used for decades despite being known to have nonidentifiable parameters.
Statistics and ComputingModel selection for probabilistic clustering using cross-validated likelihood
279 Citations2000Padhraic Smyth
The cross-validation approach, as well as penalized likelihood and McLachlan's bootstrap method, are applied to two data sets and the results from all three methods are in close agreement.
Latent Markov Models for Longitudinal Data
182 Citations2012Francesco Bartolucci, Alessio Farcomeni +1 more
This book discusses Latent Markov Modeling as a guide to Bayesian inference via reversible jump, and its applications include selection and hypothesis testing, and modeling and inference of latent variable models and their applications.
Journal of the American Statistical AssociationA Multivariate Extension of the Dynamic Logit Model for Longitudinal Data Based on a Latent Markov Heterogeneity Structure
132 Citations2009Francesco Bartolucci, Alessio Farcomeni
Computational Statistics & Data AnalysisDirect maximization of the likelihood of a hidden Markov model
46 Citations2008Rolf Turner
It is found that a straightforward implementation of Newton's method sometimes works but is unreliable, and a form of the Levenberg-Marquardt algorithm appears to provide excellent reliability.
Structural Equation Modeling A Multidisciplinary JournalAvoiding and Correcting Bias in Score-Based Latent Variable Regression With Discrete Manifest Items
32 Citations2008Irene R. R. Lu, David R. Thomas
The journal of nutrition health & agingHealth care for older people in Italy: The U.L.I.S.S.E. project (Un Link Informatico sui Servizi Sanitari Esistenti per l'anziano — a computerized network on health care services for older people)
31 Citations2010Fabrizia Lattanzio, Chiara Mussi +11 more
Health care services for older people in Italy are currently inadequate to manage the complexity of the older patients, and an important effort should be undertaken to create a more integrated health care system.
Computational Statistics & Data AnalysisA simple and fast alternative to the EM algorithm for incomplete categorical data and latent class models
22 Citations2001Andrzej T. Gałecki, Rth Ho +1 more
A faster alternative to these EM approaches that improves upon a recently introduced maximum likelihood-based alternative by Molenberghs and Goetghebeur by employing a less tedious iteratively reweighted least-squares (IRLS) approach than the Newton–Raphson procedure used by MG.
Statistics in MedicineFitting and interpreting continuous‐time latent Markov models for panel data
22 Citations2013Jane Lange, Vladimir N. Minin
A robust and efficient expectation–maximization algorithm is developed that outperforms alternative methods measured in terms of time to convergence and robustness, and illustrates the interpretive power of latent CTMC models for describing disease processes on a dataset of lung transplant patients.
Computational Statistics & Data AnalysisMaximum likelihood estimation of an extended latent Markov model for clustered binary panel data
8 Citations2006Francesco Bartolucci, Valentina Nigro