Finite mixture varying coefficient models for analyzing longitudinal heterogenous data

This paper aims to develop a mixture model to study heterogeneous longitudinal data on the treatment effect of heroin use from a California Civil Addict Program. Each component of the mixture is characterized by a varying coefficient mixed effect model. We use the Bayesian P-splines approach to approximate the varying coefficient functions. We develop Markov chain Monte Carlo algorithms to estimate the smooth functions, unknown parameters, and latent variables in the model. We use modified deviance information criterion to determine the number of components in the mixture. A simulation study demonstrates that the modified deviance information criterion selects the correct number of components and the estimation of unknown quantities is accurate. We apply the proposed model to the heroin treatment study. Furthermore, we identify heterogeneous longitudinal patterns.

[1]  Xin-Yuan Song,et al.  Bayesian model selection for mixtures of structural equation models with an unknown number of components. , 2003, The British journal of mathematical and statistical psychology.

[2]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[3]  E. V. van Ameijden,et al.  Long-term outcome of chronic drug use: the Amsterdam Cohort Study among Drug Users. , 2005, American journal of epidemiology.

[4]  Scott M. Berry,et al.  Bayesian Smoothing and Regression Splines for Measurement Error Problems , 2002 .

[5]  C. Robert,et al.  Computational and Inferential Difficulties with Mixture Posterior Distributions , 2000 .

[6]  Ajay Jasra,et al.  Markov Chain Monte Carlo Methods and the Label Switching Problem in Bayesian Mixture Modeling , 2005 .

[7]  M. Plummer.  Penalized loss functions for Bayesian model comparison. , 2008, Biostatistics.

[8]  Wenxuan Zhong,et al.  Penalized Clustering of Large-Scale Functional Data With Multiple Covariates , 2008, 0801.2555.

[9]  Anastasios Panagiotelis,et al.  Bayesian identification, selection and estimation of semiparametric functions in high-dimensional additive models , 2008 .

[10]  Sik-Yum Lee,et al.  Structural equation modelling: A Bayesian approach. , 2007 .

[11]  Chunfeng Huang,et al.  Smoothing spline estimation in varying‐coefficient models , 2004 .

[12]  S. Lang,et al.  Bayesian P-Splines , 2004 .

[13]  Jianhua Z. Huang,et al.  Varying‐coefficient models and basis function approximations for the analysis of repeated measurements , 2002 .

[14]  P. Green.  Reversible jump Markov chain Monte Carlo computation and Bayesian model determination , 1995 .

[15]  R. Redner,et al.  Mixture densities, maximum likelihood, and the EM algorithm , 1984 .

[16]  Geoffrey J. McLachlan,et al.  Finite Mixture Models , 2019, Annual Review of Statistics and Its Application.

[17]  L. Fahrmeir,et al.  Geoadditive Survival Models , 2006 .

[18]  David Huang,et al.  Trajectories of Heroin Addiction , 2007, Evaluation review.

[19]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  W. McCarthy,et al.  Health Conditions Among Aging Narcotics Addicts: Medical Examination Results , 2004, Journal of Behavioral Medicine.

[21]  P. Green,et al.  Corrigendum: On Bayesian analysis of mixtures with an unknown number of components , 1997 .

[22]  Daniel L Gillen,et al.  The separation of timescales in Bayesian survival modeling of the time-varying effect of a time-dependent exposure. , 2008, Biostatistics.

[23]  C. Robert,et al.  Deviance information criteria for missing data models , 2006 .

[24]  S. Frühwirth-Schnatter.  Markov chain Monte Carlo Estimation of Classical and Dynamic Switching and Mixture Models , 2001 .

[25]  Joseph W Hogan,et al.  Mixtures of Varying Coefficient Models for Longitudinal Data with Discrete or Continuous Nonignorable Dropout , 2004, Biometrics.

[26]  Jianqing Fan,et al.  Statistical Estimation in Varying-Coefficient Models , 1999 .

[27]  Xin-Yuan Song,et al.  A mixture of generalized latent variable models for mixed mode and heterogeneous data , 2011, Comput. Stat. Data Anal..

[28]  M J Daniels,et al.  Dynamic conditionally linear mixed models for longitudinal data. , 2002, Biometrics.

[29]  David B Dunson,et al.  Bayesian Modeling of Time‐Varying and Waning Exposure Effects , 2003, Biometrics.

[30]  Chin-Tsang Chiang,et al.  Smoothing Spline Estimation for Varying Coefficient Models With Repeatedly Measured Dependent Variables , 2001 .

[31]  C. F. Sirmans,et al.  Spatial Modeling With Spatially Varying Coefficient Processes , 2003, Journal of the American Statistical Association.

[32]  Joseph W. Hogan,et al.  Varying-coefficient models for longitudinal processes with continuous-time informative dropout , 2009, Biostatistics.

[33]  Ludwig Fahrmeir,et al.  Bayesian varying-coefficient models using adaptive regression splines , 2001 .

[34]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[35]  M. Stephens.  Dealing with label switching in mixture models , 2000 .

[36]  Xinyuan Song,et al.  Semiparametric Latent Variable Models With Bayesian P-Splines , 2010 .

[37]  Li Ping Yang,et al.  Nonparametric smoothing estimates of time-varying coefficient models with longitudinal data , 1998 .

[38]  R. Tibshirani,et al.  Varying‐Coefficient Models , 1993 .

[39]  Bradley P. Carlin,et al.  Bayesian measures of model complexity and fit , 2002 .

[40]  P. Green,et al.  On Bayesian Analysis of Mixtures with an Unknown Number of Components (with discussion) , 1997 .

[41]  Xin-Yuan Song,et al.  A Bayesian analysis of mixture structural equation models with non‐ignorable missing responses and covariates , 2010, Statistics in medicine.