Results 1  10
of
53
Bayesian Model Assessment In Factor Analysis
, 2004
"... Factor analysis has been one of the most powerful and flexible tools for assessment of multivariate dependence and codependence. Loosely speaking, it could be argued that the origin of its success rests in its very exploratory nature, where various kinds of datarelationships amongst the variable ..."
Abstract

Cited by 104 (10 self)
 Add to MetaCart
Factor analysis has been one of the most powerful and flexible tools for assessment of multivariate dependence and codependence. Loosely speaking, it could be argued that the origin of its success rests in its very exploratory nature, where various kinds of datarelationships amongst the variables at study can be iteratively verified and/or refuted. Bayesian inference in factor analytic models has received renewed attention in recent years, partly due to computational advances but also partly to applied focuses generating factor structures as exemplified by recent work in financial time series modeling. The focus of our current work is on exploring questions of uncertainty about the number of latent factors in a multivariate factor model, combined with methodological and computational issues of model specification and model fitting. We explore reversible jump MCMC methods that build on sets of parallel Gibbs samplingbased analyses to generate suitable empirical proposal distributions and that address the challenging problem of finding e#cient proposals in highdimensional models. Alternative MCMC methods based on bridge sampling are discussed, and these fully Bayesian MCMC approaches are compared with a collection of popular model selection methods in empirical studies.
Sparse Bayesian infinite factor models
"... We focus on sparse modeling of highdimensional covariance matrices using Bayesian latent factor models. We propose a multiplicative gamma process shrinkage prior on the factor loadings which allows introduction of infinitely many factors, with the loadings increasingly shrunk toward zero as the col ..."
Abstract

Cited by 52 (16 self)
 Add to MetaCart
(Show Context)
We focus on sparse modeling of highdimensional covariance matrices using Bayesian latent factor models. We propose a multiplicative gamma process shrinkage prior on the factor loadings which allows introduction of infinitely many factors, with the loadings increasingly shrunk toward zero as the column index increases. We use our prior on a parameter expanded loadings matrix to avoid the order dependence typical in factor analysis models and develop a highly efficient Gibbs sampler that scales well as data dimensionality increases. The gain in efficiency is achieved by the joint conjugacy property of the proposed prior, which allows block updating of the loadings matrix. We propose an adaptive Gibbs sampler for automatically truncating the infinite loadings matrix through selection of the number of important factors. Theoretical results are provided on the support of the prior and truncation approximation bounds. A fast algorithm is proposed to produce approximate Bayes estimates. Latent factor regression methods are developed for prediction and variable selection in applications with highdimensional correlated predictors. Operating characteristics are assessed through simulation studies and the approach is applied to predict survival after chemotherapy from gene expression data.
Default prior distributions and efficient posterior computation in Bayesian factor analysis
 Journal of Computational and Graphical Statistics
, 2009
"... Factor analytic models are widely used in social sciences. These models have also proven useful for sparse modeling of the covariance structure in multidimensional data. Normal prior distributions for factor loadings and inverse gamma prior distributions for residual variances are a popular choice b ..."
Abstract

Cited by 28 (6 self)
 Add to MetaCart
(Show Context)
Factor analytic models are widely used in social sciences. These models have also proven useful for sparse modeling of the covariance structure in multidimensional data. Normal prior distributions for factor loadings and inverse gamma prior distributions for residual variances are a popular choice because of their conditionally conjugate form. However, such prior distributions require elicitation of many hyperparameters and tend to result in poorly behaved Gibbs samplers. In addition, one must choose an informative specification, as high variance prior distributions face problems due to impropriety of the posterior distribution. This article proposes a default, heavy tailed prior distribution specification, which is induced through parameter expansion while facilitating efficient posterior computation. We also develop an approach to allow uncertainty in the number of factors. The methods are illustrated through simulated examples and epidemiology and toxicology applications.
Bayesian analysis using Mplus: Technical implementation
, 2010
"... In this note we describe the implementation details for estimating latent variable models with the Bayesian estimator in Mplus. The algorithm used in Mplus is Markov Chain Monte Carlo (MCMC) based on the Gibbs sampler, see Gelman et al. (2004). ..."
Abstract

Cited by 16 (8 self)
 Add to MetaCart
In this note we describe the implementation details for estimating latent variable models with the Bayesian estimator in Mplus. The algorithm used in Mplus is Markov Chain Monte Carlo (MCMC) based on the Gibbs sampler, see Gelman et al. (2004).
Commentary: practical advantages of Bayesian analysis of epidemiologic data
 Am J Epidemiol
, 2001
"... In the past decade, there have been enormous advances in the use of Bayesian methodology for analysis of epidemiologic data, and there are now many practical advantages to the Bayesian approach. Bayesian models can easily accommodate unobserved variables such as an individual’s true disease status i ..."
Abstract

Cited by 15 (0 self)
 Add to MetaCart
(Show Context)
In the past decade, there have been enormous advances in the use of Bayesian methodology for analysis of epidemiologic data, and there are now many practical advantages to the Bayesian approach. Bayesian models can easily accommodate unobserved variables such as an individual’s true disease status in the presence of diagnostic error. The use of prior probability distributions represents a powerful mechanism for incorporating information from previous studies and for controlling confounding. Posterior probabilities can be used as easily interpretable alternatives to p values. Recent developments in Markov chain Monte Carlo methodology facilitate the implementation of Bayesian analyses of complex data sets containing missing observations and multidimensional outcomes.Tools are now available that allow epidemiologists to take advantage of this powerful approach to assessment of exposuredisease relations. Am J Epidemiol 2001;153:1222–6. Bayes theorem; epidemiologic methods; hierarchical Bayes; latent variable; Markov chain Monte Carlo; posterior probability; prior distribution Received for publication December 26, 2000, and accepted for publication March 12, 2001.
Quasi Maximum Likelihood Estimation of Structural Equation Models With Multiple Interaction and Quadratic Effects
"... The development of statistically efficient and computationally practicable estimation methods for the analysis of structural equation models with multiple nonlinear effects has been called for by substantive researchers in psychology, marketing research, and sociology. But the development of efficie ..."
Abstract

Cited by 14 (0 self)
 Add to MetaCart
The development of statistically efficient and computationally practicable estimation methods for the analysis of structural equation models with multiple nonlinear effects has been called for by substantive researchers in psychology, marketing research, and sociology. But the development of efficient methods is complicated by the fact that a nonlinear model structure implies specifically nonnormal multivariate distributions for the indicator variables. In this paper, nonlinear structural equation models with quadratic forms are introduced and a new QuasiMaximum Likelihood method for simultaneous estimation of model parameters is developed with the focus on statistical efficiency and computational practicability. The QuasiML method is based on an approximation of the nonnormal density function of the joint indicator vector by a product of a normal and a conditionally normal density. The results of MonteCarlo studies for the new QuasiML method indicate that the parameter estimation is almost as efficient as ML estimation, whereas ML estimation is only computationally practical for elementary models. Also, the QuasiML method outperforms other currently available methods with respect to efficiency. It is demonstrated in a MonteCarlo study that the QuasiML method permits computationally feasible and very efficient analysis of models with multiple latent nonlinear effects. Finally, the applicability of the QuasiML method is illustrated by an empirical example of an aging study in psychology. Key words: structural equation modeling, quadratic form of normal variates, latent interaction effect, moderator effect, QuasiML estimation, variance function model. 1 1.
A Bayesian semiparametric latent variable model for binary, ordinal and continuous response. Dissertation
, 2005
"... In this article we introduce a latent variable model (LVM) for mixed ordinal and continuous responses, where covariate effects on the continuous latent variables are modelled through a flexible semiparametric predictor. We extend existing LVM with simple linear covariate effects by including nonpara ..."
Abstract

Cited by 11 (2 self)
 Add to MetaCart
In this article we introduce a latent variable model (LVM) for mixed ordinal and continuous responses, where covariate effects on the continuous latent variables are modelled through a flexible semiparametric predictor. We extend existing LVM with simple linear covariate effects by including nonparametric components for nonlinear effects of continuous covariates and interactions with other covariates as well as spatial effects. Full Bayesian modelling is based on penalized spline and Markov random field priors and is performed by computationally efficient Markov chain Monte Carlo (MCMC) methods. We apply our approach to a large German social science survey which motivated our methodological development.
Latent variable modelling: A survey
 Scandinavian Journal of Statistics
"... ABSTRACT. Latent variable modelling has gradually become an integral part of mainstream statistics and is currently used for a multitude of applications in different subject areas. Examples of ‘traditional ’ latent variable models include latent class models, item–response models, common factor mode ..."
Abstract

Cited by 8 (2 self)
 Add to MetaCart
ABSTRACT. Latent variable modelling has gradually become an integral part of mainstream statistics and is currently used for a multitude of applications in different subject areas. Examples of ‘traditional ’ latent variable models include latent class models, item–response models, common factor models, structural equation models, mixed or random effects models and covariate measurement error models. Although latent variables have widely different interpretations in different settings, the models have a very similar mathematical structure. This has been the impetus for the formulation of general modelling frameworks which accommodate a wide range of models. Recent developments include multilevel structural equation models with both continuous and discrete latent variables, multiprocess models and nonlinear latent variable models.
POSTERIOR PREDICTIVE MODEL CHECKING FOR MULTIDIMENSIONALITY IN ITEM RESPONSE THEORY AND BAYESIAN NETWORKS
, 2006
"... If data exhibit a dimensional structure more complex than what is assumed, key conditional independence assumptions of the hypothesized model do not hold. The current work pursues posterior predictive model checking, a flexible family of Bayesian model checking procedures, as a tool for criticizing ..."
Abstract

Cited by 8 (3 self)
 Add to MetaCart
If data exhibit a dimensional structure more complex than what is assumed, key conditional independence assumptions of the hypothesized model do not hold. The current work pursues posterior predictive model checking, a flexible family of Bayesian model checking procedures, as a tool for criticizing models in light of inadequately modeled dimensional structure. Factors hypothesized to influence dimensionality and dimensionality assessment are couched in conditional covariance theory and conveyed via geometric representations of multidimensionality. These factors and their hypothesized effects motivate a simulation study that investigates posterior predictive model checking in the context of item response theory for dichotomous observables. A unidimensional model is fit to data that follow compensatory or conjunctive multidimensional item response models to assess the utility of conducting posterior predictive model checking. Discrepancy measures are formulated at the level of individual items and pairs of items. A second study draws from the results of the first study and investigates the model checking techniques in the context of multidimensional Bayesian networks with inhibitory effects. Key findings include support for the