Bayesian semiparametric analysis of structural equation models with mixed continuous and unordered categorical variables
Statistics in MedicinePublished 26 May 2009
Xinyuan Song, Ye‐Mao Xia, Sik‐Yum Lee
Citations26
SJR quartileQ1
SJR score1.27
SNIP1.33
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 Bayesian semparametric SEM with covariates, and mixed continuous and unordered categorical variables, in which the explanatory latent variables in the structural equation are modeled via an appropriate truncated Dirichlet process with a stick‐breaking procedure.
Abstract
Recently, structural equation models (SEMs) have been applied for analyzing interrelationships among observed and latent variables in biological and medical research. Latent variables in these models are typically assumed to have a normal distribution. This article considers a Bayesian semparametric SEM with covariates, and mixed continuous and unordered categorical variables, in which the explanatory latent variables in the structural equation are modeled via an appropriate truncated Dirichlet process with a stick-breaking procedure. Results obtained from a simulation study and an analysis of a real medical data set are presented to illustrate the methodology.
Keywords
Computer ScienceMathematics
The Journal of Chemical PhysicsEquation of State Calculations by Fast Computing Machines
37,074 Citations1953N. Metropolis, Arianna W. Rosenbluth +3 more
BiometrikaMonte Carlo sampling methods using Markov chains and their applications
15,200 Citations1970W. Keith Hastings
Journal of the American Statistical AssociationBayes Factors
12,293 Citations1995Robert E. Kass, Adrian E. Raftery
Journal of the Royal Statistical Society Series D (The Statistician)Structural Equation Modeling: Concepts, Issues, and Applications.
6,867 Citations1996Murray Smith, Rick H. Hoyle
The Annals of StatisticsA Bayesian Analysis of Some Nonparametric Problems
4,812 Citations1973Thomas S. Ferguson
A Constructive Definition of Dirichlet Priors
1,611 Citations1991Jayaram Sethuraman
Journal of the American Statistical AssociationGibbs Sampling Methods for Stick-Breaking Priors
1,556 Citations2001Hemant Ishwaran, Lancelot F. James
Two general types of Gibbs samplers that can be used to fit posteriors of Bayesian hierarchical models based on stick-breaking priors are presented and the blocked Gibbs sampler, based on an entirely different approach that works by directly sampling values from the posterior of the random measure.
Psychological BulletinCan test statistics in covariance structure analysis be trusted?
1,360 Citations1992Li‐tze Hu, Peter M. Bentler +1 more
Covariance structure analysis uses chi 2 goodness-of-fit test statistics whose adequacy is not known to evaluate the behavior of 6 test statistics, which perform dramatically differently under 7 distributional conditions at 6 sample sizes.
New England Journal of MedicineReversal of Lesions of Diabetic Nephropathy after Pancreas Transplantation
1,128 Citations1998Paola Fioretto, Michael W. Steffes +3 more
New England Journal of MedicineFamilial Clustering of Diabetic Kidney Disease
995 Citations1989Elizabeth R. Seaquist, Frederick C. Goetz +2 more
Logistic regression analysis found nephropathy in the proband to be the only factor significantly predictive of the renal status of the diabetic sibling, consistent with the hypothesis that heredity helps to determine susceptibility to diabetic nephopathy.
Statistical ScienceSimulating normalizing constants: from importance sampling to bridge sampling to path sampling
974 Citations1998Andrew Gelman, Xiao‐Li Meng
It is shown that the acceptance ratio method and thermodynamic integration are natural generalizations of importance sampling, which is most familiar to statistical audiences.
PsychometrikaGeneralized Multilevel Structural Equation Modeling
826 Citations2004Sophia Rabe‐Hesketh, Anders Skrondal +1 more
Maximum likelihood estimation and empirical Bayes latent score prediction within the GLLAMM framework can be performed using adaptive quadrature in gllamm, a freely available program running in Stata.
Archives of Internal MedicineMain Risk Factors for Nephropathy in Type 2 Diabetes Mellitus Are Plasma Cholesterol Levels, Mean Blood Pressure, and Hyperglycemia
451 Citations1998Mordchai Ravid, David Brosh +3 more
The combination of blood pressure values in the high-normal range with moderately elevated levels of total cholesterol and hemoglobin A1c defines a high-risk group for the progression to diabetic nephropathy and for clinical events related to arteriosclerotic cardiovascular disease.
Journal of the Royal Statistical Society Series B (Statistical Methodology)Predictive Model Selection
358 Citations1995Purushottam W. Laud, Joseph G. Ibrahim
Journal of EconometricsA Bayesian analysis of the multinomial probit model using marginal data augmentation
224 Citations2004Kosuke Imai, David A. van Dyk
A set of new Markov chain Monte Carlo algorithms for Bayesian analysis of the multinomial probit model are introduced, which are as quick to converge as the fastest methods but with a more attractive prior.
Journal of the American Statistical AssociationBayesian Semiparametric Median Regression Modeling
202 Citations2001Athanasios Kottas, Alan E. Gelfand
Statistical Methods in Medical ResearchStructural equation models in medical research
200 Citations1992P. M. Bentler, JA Stein
The underlying statistical theory and rationale of SEM, both confirmatory factor analysis and latent variable path models are discussed, and the applicability of SEM to assessment of reliability and validity is noted.
BiometricsA Semiparametric Bayesian Approach to the Random Effects Model
170 Citations1998Ken Kleinman, Joseph G. Ibrahim
A Bayesian model is provided that allows the random effects to have a nonparametric prior distribution in longitudinal random effects models and a Dirichlet process prior is proposed for the distribution of therandom effects.
Journal of the American Statistical AssociationMarginal Likelihood and Bayes Factors for Dirichlet Process Mixture Models
140 Citations2003Sanjib Basu, Siddhartha Chib
A method for comparingSemiparametric Bayesian models, constructed under the Dirichlet process mixture (DPM) framework, with alternative semiparameteric or parametericBayesian models is presented, apparently the first method of its kind.
Journal of the American Statistical AssociationBayesian Model Selection in Finite Mixtures by Marginal Density Decompositions
128 Citations2001Hemant Ishwaran, Lancelot F. James +1 more
A weighted Bayes factor method for consistently estimating d that can be implemented by an iid generalized weighted Chinese restaurant (GWCR) Monte Carlo algorithm and the performance of the new GWCR model selection procedure is compared with that of the Akaike information criterion and the Bayes information criterion implemented through an EM algorithm.
Journal of the American Society of NephrologyGenetics of diabetic nephropathy.
123 Citations1996Hans‐Henrik Parving, Lise Tarnow +1 more
A meta-analysis does not support the suggestion that this factor plays any role for the initiation of diabetic nephropathy, but studies in diabetic and non-diabetic glomerulopathies have clearly demonstrated a deleterious effect of the deletion polymorphism in the angiotensin-converting enzyme on the progression of kidney function.
Journal of the American Statistical AssociationAnalysis of Covariance Structures under Elliptical Distributions
109 Citations1987Alexander Shapiro, Michael W. Browne
Journal of the American Statistical AssociationA Bayesian Population Model with Hierarchical Mixture Priors Applied to Blood Count Data
103 Citations1997Peter Müller, Gary L. Rosner
A class of nonlinear population models with nonparametric second-stage priors with a flexible class of mixtures are proposed for analyzing nonlinear growth curves fit to multiple measurements from study subjects.
Journal of the American Statistical AssociationA Predictive Approach to the Analysis of Designed Experiments
96 Citations1994Joseph G. Ibrahim, Purushottam W. Laud
Viewing the analysis of designed experiments as a model selection problem, the use of a predictive Bayesian criterion based on the predictive density of a replicate experiment (PDRE) is introduced and compared with the usual F statistic for two nested models.
Criterion-based methods for Bayesian model assessment
79 Citations2001Joseph G. Ibrahim, Ming Hui Chen +1 more
A general Bayesian criterion for model assessment, motivated by earlier work of Ibrahim and Laud (1994) and related to a criterion of Gelfand and Ghosh (1998), and a calibration of the L measure, defined as the prior predictive distribution of the difference between the L measures of the candidate model and the criterion minimizing model are proposed.
Computational Statistics & Data AnalysisSemiparametric Bayes hierarchical models with mean and variance constraints
61 Citations2010Mingan Yang, David B. Dunson +1 more
A centered stick-breaking process (CSBP) is proposed, which induces mean and variance constraints on an unknown distribution in a hierarchical model by viewing an unconstrained stick- Breaking process as a parameter-expanded version of a CSBP.
BiometricsMaximum Likelihood Analysis of a General Latent Variable Model with Hierarchically Mixed Data
58 Citations2004Sik‐Yum Lee, Xin‐Yuan Song
A general two-level latent variable model is developed to provide a comprehensive framework for model comparison of various submodels and can be applied to hierarchically mixed continuous, dichotomous, and polytomous data.
Diabetes CarePrognostic Effect of Insertion/Deletion Polymorphism of the ACE Gene on Renal and Cardiovascular Clinical Outcomes in Chinese Patients With Type 2 Diabetes
53 Citations2005Ying Wang, Maggie Ng +6 more
The DD genotype of the ACE I/D polymorphism was an independent risk factor for renal but not cardiovascular end points in Chinese patients with type 2 diabetes.
Journal of the American Statistical AssociationStatistical Inference Based on Pseudo-Maximum Likelihood Estimators in Elliptical Populations
53 Citations1993Yutaka Kano, Maia Berkane +1 more
The Annals of StatisticsBayesian Nonparametric Estimation of the Median; Part I: Computation of the Estimates
49 Citations1985Hani Doss
DiabetologiaCommon variants in the β2-(Gln27Glu) and β3-(Trp64Arg) -adrenoceptor genes are associated with elevated serum NEFA concentrations and Type II diabetes
47 Citations2001Martin Carlsson, Marju Orho‐Melander +2 more
Common variants in the beta2- and beta3- adrenoceptor genes are associated with increased fasting insulin and NEFA concentrations and could increase susceptibility to Type II diabetes.
Statistics in MedicineSemiparametric Bayesian analysis of structural equation models with fixed covariates
45 Citations2007Sik‐Yum Lee, Bin Lu +1 more
A general Bayesian framework is provided in which a semiparametric hierarchical modeling with an approximate truncation Dirichlet process prior distribution is specified for the latent variables in SEMs with covariates.
Statistics in MedicineBayesian analysis of structural equation models with multinomial variables and an application to type 2 diabetic nephropathy
26 Citations2006Xin‐Yuan Song, Sik‐Yum Lee +3 more
A novel nonlinear structural equation model is established for analysing mixed continuous and multinomial data that can be missing at random and applied to a case–control cohort of type 2 diabetic patients with nephropathy.
Statistics in MedicineA two‐level structural equation model approach for analyzing multivariate longitudinal responses
23 Citations2008Song Xin-yuan, Sik‐Yum Lee +1 more
A maximum likelihood (ML) approach is developed for the estimation of parameters and model comparison and results of a simulation study indicate that the performance of the ML estimation is satisfactory.