Bayesian Analysis of Multivariate Latent Curve Models With Nonlinear Longitudinal Latent Effects.

In longitudinal studies, investigators often measure multiple variables at multiple time points and are interested in investigating individual differences in patterns of change on those variables. Furthermore, in behavioral, social, psychological, and medical research, investigators often deal with latent variables that cannot be observed directly and should be measured by 2 or more manifest variables. Longitudinal latent variables occur when the corresponding manifest variables are measured at multiple time points. Our primary interests are in studying the dynamic change of longitudinal latent variables and exploring the possible interactive effect among the latent variables.Much of the existing research in longitudinal studies focuses on studying change in a single observed variable at different time points. In this article, we propose a novel latent curve model (LCM) for studying the dynamic change of multivariate manifest and latent variables and their linear and interaction relationships. The proposed LCM has the following useful features: First, it can handle multivariate variables for exploring the dynamic change of their relationships, whereas conventional LCMs usually consider change in a univariate variable. Second, it accommodates both first- and second-order latent variables and their interactions to explore how changes in latent attributes interact to produce a joint effect on the growth of an outcome variable. Third, it accommodates both continuous and ordered categorical data, and missing data.

Bayesian semiparametric analysis of structural equation models with mixed continuous and unordered categorical variables.

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.

Robust model fitting for the non linear structural equation model under normal theory.

Structural equation modelling has been widely applied in behavioural, educational, medical, social, and psychological research. The classical maximum likelihood estimate is vulnerable to outliers and non-normal data. In this paper, a robust estimation method for the nonlinear structural equation model is proposed. This method gives more weight to data that are likely to occur based on the structure of the posited model, and effectively downweights the influence of outliers. An algorithm is proposed to obtain the robust estimator. Asymptotic properties of the proposed method are investigated, which include the asymptotic distribution of the estimator, and some statistics for hypothesis testing. Results from a simulation study and a real data example show that our procedure is effective.

Non-linear structural equation models with correlated continuous and discrete data.

Structural equation models (SEMs) have been widely applied to examine interrelationships among latent and observed variables in social and psychological research. Motivated by the fact that correlated discrete variables are frequently encountered in practical applications, a non-linear SEM that accommodates covariates, and mixed continuous, ordered, and unordered categorical variables is proposed. Maximum likelihood methods for estimation and model comparison are discussed. One real-life data set about cardiovascular disease is used to illustrate the methodologies.

Longitudinal analysis of quality of life for stroke survivors using latent curve models.

BACKGROUND AND PURPOSE: For the survivors, activities of daily living, handicap, and depression have a significant impact on health-related quality of life (HRQOL). How the dynamic changes of these variables relate to HRQOL over time in the subacute phase of stroke recovery has not been investigated. The objective of this study was to study longitudinal behaviors of HRQOL of the stroke survivors in relation to the changes in activities of daily living, handicap, and depression after stroke. METHODS: This was a prospective cohort study of first disabling patients with stroke. Subjects were interviewed at 3, 6, and 12 months after stroke for modified Barthel Index, London Handicap Scale, Geriatric Depression Scale, and the World Health Organization Quality of Life questionnaire (abbreviated Hong Kong version). A latent curve model was developed to analyze how the dynamic changes in activities of daily living, handicap, and depressive mood related to the changes in HRQOL. RESULTS: Two hundred forty-seven of 303 patients (82%) followed up at 3 months after stroke could complete the quality-of-life questionnaire. Their mean age was 68.8 years. The latent curve model analysis revealed that initial physical health HRQOL was independently associated with activities of daily living, handicap, and depression. The other 3 HRQOL domain scores were primarily associated with depression only. The rates of change in all 4 domains of HRQOL were significantly and inversely associated with rate of change in the Geriatric Depression Scale only. CONCLUSIONS: Change in mood in the postacute phase of stroke recovery is the most significant determinant of change in HRQOL. More attention should be paid to the detection and management of poststroke depression.

Influence analysis for the factor analysis model with ranking data.

Influence analysis is an important component of data analysis, and the local influence approach has been widely applied to many statistical models to identify influential observations and assess minor model perturbations since the pioneering work of Cook (1986). The approach is often adopted to develop influence analysis procedures for factor analysis models with ranking data. However, as this well-known approach is based on the observed data likelihood, which involves multidimensional integrals, directly applying it to develop influence analysis procedures for the factor analysis models with ranking data is difficult. To address this difficulty, a Monte Carlo expectation and maximization algorithm (MCEM) is used to obtain the maximum-likelihood estimate of the model parameters, and measures for influence analysis on the basis of the conditional expectation of the complete data log likelihood at the E-step of the MCEM algorithm are then obtained. Very little additional computation is needed to compute the influence measures, because it is possible to make use of the by-products of the estimation procedure. Influence measures that are based on several typical perturbation schemes are discussed in detail, and the proposed method is illustrated with two real examples and an artificial example.

A two-level structural equation model approach for analyzing multivariate longitudinal responses.

The analysis of longitudinal data to study changes in variables measured repeatedly over time has received considerable attention in many fields. This paper proposes a two-level structural equation model for analyzing multivariate longitudinal responses that are mixed continuous and ordered categorical variables. The first-level model is defined for measures taken at each time point nested within individuals for investigating their characteristics that are changed with time. The second level is defined for individuals to assess their characteristics that are invariant with time. The proposed model accommodates fixed covariates, nonlinear terms of the latent variables, and missing data. A maximum likelihood (ML) approach is developed for the estimation of parameters and model comparison. Results of a simulation study indicate that the performance of the ML estimation is satisfactory. The proposed methodology is applied to a longitudinal study concerning cocaine use.

Semiparametric Bayesian analysis of structural equation models with fixed covariates.

Latent variables play the most important role in structural equation modeling. In almost all existing structural equation models (SEMs), it is assumed that the distribution of the latent variables is normal. As this assumption is likely to be violated in many biomedical researches, a semiparametric Bayesian approach for relaxing it is developed in this paper. In the context of SEMs with covariates, we provide a general Bayesian framework in which a semiparametric hierarchical modeling with an approximate truncation Dirichlet process prior distribution is specified for the latent variables. The stick-breaking prior and the blocked Gibbs sampler are used for efficient simulation in the posterior analysis. The developed methodology is applied to a study of kidney disease in diabetes patients. A simulation study is conducted to reveal the empirical performance of the proposed approach. Supplementary electronic material for this paper is available in Wiley InterScience at http://www.mrw.interscience.wiley.com/suppmat/1097-0258/suppmat/.

Bayesian analysis of latent variable models with non-ignorable missing outcomes from exponential family.

To provide a comprehensive framework for analysing complex non-normal medical and biological data, we propose a Bayesian approach for a non-linear latent variable model with covariates, and non-ignorable missing data, under the exponential family of distributions. The non-ignorable missing mechanism is defined via a logistic regression model. Based on conjugate prior distributions, full conditional distributions for the implementation of Markov chain Monte Carlo methods in simulating observations from the joint posterior distribution are derived. These observations are used in computing the Bayesian estimates, as well as in implementing a path sampling procedure to evaluate the Bayes factor for model comparison. The proposed methods are illustrated using real data from a study on the non-adherence of hypertension patients.

Bayesian analysis of structural equation models with multinomial variables and an application to type 2 diabetic nephropathy.

There is now increasing evidence proving that many complex diseases can be significantly influenced by correlated phenotype and genotype variables, as well as their interactions. Effective and rigorous assessment of such influence is difficult, because the number of phenotype and genotype variables of interest may not be small, and a genotype variable is an unordered categorical variable that follows a multinomial distribution. To address the problem, we establish a novel nonlinear structural equation model for analysing mixed continuous and multinomial data that can be missing at random. A confirmatory factor analysis model with Kronecker product is proposed for grouping the manifest continuous and multinomial variables into latent variables according to their functions; and a nonlinear structural equation is formulated to assess the linear and interaction effects of the independent latent variables to the dependent latent variables. Bayesian methods for estimation and model comparison are developed through Markov chain Monte Carlo techniques and path sampling. The newly developed methodologies are applied to a case-control cohort of type 2 diabetic patients with nephropathy.

Bayesian analysis of structural equation models with mixed exponential family and ordered categorical data.

Structural equation models are very popular for studying relationships among observed and latent variables. However, the existing theory and computer packages are developed mainly under the assumption of normality, and hence cannot be satisfactorily applied to non-normal and ordered categorical data that are common in behavioural, social and psychological research. In this paper, we develop a Bayesian approach to the analysis of structural equation models in which the manifest variables are ordered categorical and/or from an exponential family. In this framework, models with a mixture of binomial, ordered categorical and normal variables can be analysed. Bayesian estimates of the unknown parameters are obtained by a computational procedure that combines the Gibbs sampler and the Metropolis-Hastings algorithm. Some goodness-of-fit statistics are proposed to evaluate the fit of the posited model. The methodology is illustrated by results obtained from a simulation study and analysis of a real data set about non-adherence of hypertension patients in a medical treatment scheme.

Model comparison of generalized linear mixed models.

Generalized linear mixed models (GLMMs) have been widely appreciated in biological and medical research. Maximum likelihood estimation has received a great deal of attention. Comparatively, not much has been done on model comparison or hypotheses testing. In this article, we propose a path sampling procedure to compute the observed-data log-likelihood function, so that the Bayesian information criterion (BIC) can be applied to model comparison or hypothesis testing. Advantages of the proposed path sampling procedure are discussed. Two medical data sets are analysed for providing illustrative examples of the proposed methodology.

Bayesian Analysis of Structural Equation Models With Nonlinear Covariates and Latent Variables.

In this article, we formulate a nonlinear structural equation model (SEM) that can accommodate covariates in the measurement equation and nonlinear terms of covariates and exogenous latent variables in the structural equation. The covariates can come from continuous or discrete distributions. A Bayesian approach is developed to analyze the proposed model. Markov chain Monte Carlo methods for obtaining Bayesian estimates and their standard error estimates, highest posterior density intervals, and a PP p value are developed. Results obtained from two simulation studies are reported to respectively reveal the empirical performance of the proposed Bayesian estimation in analyzing complex nonlinear SEMs, and in analyzing nonlinear SEMs with the normal assumption of the exogenous latent variables violated. The proposed methodology is further illustrated by a real example. Detailed interpretation about the interaction terms is presented.

Maximum Likelihood Analysis of Nonlinear Structural Equation Models With Dichotomous Variables.

In this article, a maximum likelihood approach is developed to analyze structural equation models with dichotomous variables that are common in behavioral, psychological and social research. To assess nonlinear causal effects among the latent variables, the structural equation in the model is defined by a nonlinear function. The basic idea of the development is to augment the observed dichotomous data with the hypothetical missing data that involve the latent underlying continuous measurements and the latent variables in the model. An EM algorithm is implemented. The conditional expectation in the E-step is approximated via observations simulated from the appropriate conditional distributions by a Metropolis-Hastings algorithm within the Gibbs sampler, whilst the M-step is completed by conditional maximization. Convergence is monitored by bridge sampling. Standard errors are also obtained. Results from a simulation study and a real example are presented to illustrate the methodology.

Maximum likelihood analysis of a general latent variable model with hierarchically mixed data.

A general two-level latent variable model is developed to provide a comprehensive framework for model comparison of various submodels. Nonlinear relationships among the latent variables in the structural equations at both levels, as well as the effects of fixed covariates in the measurement and structural equations at both levels, can be analyzed within the framework. Moreover, the methodology can be applied to hierarchically mixed continuous, dichotomous, and polytomous data. A Monte Carlo EM algorithm is implemented to produce the maximum likelihood estimate. The E-step is completed by approximating the conditional expectations through observations that are simulated by Markov chain Monte Carlo methods, while the M-step is completed by conditional maximization. A procedure is proposed for computing the complicated observed-data log likelihood and the BIC for model comparison. The methods are illustrated by using a real data set.

Bayesian analysis of two-level nonlinear structural equation models with continuous and polytomous data.

Two-level structural equation models with mixed continuous and polytomous data and nonlinear structural equations at both the between-groups and within-groups levels are important but difficult to deal with. A Bayesian approach is developed for analysing this kind of model. A Markov chain Monte Carlo procedure based on the Gibbs sampler and the Metropolis-Hasting algorithm is proposed for producing joint Bayesian estimates of the thresholds, structural parameters and latent variables at both levels. Standard errors and highest posterior density intervals are also computed. A procedure for computing Bayes factor, based on the key idea of path sampling, is established for model comparison.

Bayesian model comparison of nonlinear structural equation models with missing continuous and ordinal categorical data.

Missing data are very common in behavioural and psychological research. In this paper, we develop a Bayesian approach in the context of a general nonlinear structural equation model with missing continuous and ordinal categorical data. In the development, the missing data are treated as latent quantities, and provision for the incompleteness of the data is made by a hybrid algorithm that combines the Gibbs sampler and the Metropolis-Hastings algorithm. We show by means of a simulation study that the Bayesian estimates are accurate. A Bayesian model comparison procedure based on the Bayes factor and path sampling is proposed. The required observations from the posterior distribution for computing the Bayes factor are simulated by the hybrid algorithm in Bayesian estimation. Our simulation results indicate that the correct model is selected more frequently when the incomplete records are used in the analysis than when they are ignored. The methodology is further illustrated with a real data set from a study concerned with an AIDS preventative intervention for Filipina sex workers.

Evaluation of the Bayesian and Maximum Likelihood Approaches in Analyzing Structural Equation Models with Small Sample Sizes.

The main objective of this article is to investigate the empirical performances of the Bayesian approach in analyzing structural equation models with small sample sizes. The traditional maximum likelihood (ML) is also included for comparison. In the context of a confirmatory factor analysis model and a structural equation model, simulation studies are conducted with the different magnitudes of parameters and sample sizes n = da, where d = 2, 3, 4 and 5, and a is the number of unknown parameters. The performances are evaluated in terms of the goodness-of-fit statistics, and various measures on the accuracy of the estimates. The conclusion is: for data that are normally distributed, the Bayesian approach can be used with small sample sizes, whilst ML cannot.

Comparison of Approaches in Estimating Interaction and Quadratic Effects of Latent Variables.

Various approaches using the maximum likelihood (ML) option of the LISREL program and products of indicators have been proposed to analyze structural equation models with non-linear latent effects on the basis of Kenny and Judd's formulation. Recently, some methods based on the Bayesian approach and the exact ML approaches have been developed. This article reviews, elaborates and compares several approaches for analyzing nonlinear models with interaction and/or quadratic effects. A total of four approaches are examined, including the product indicator ML approaches proposed by Jaccard and Wan (1995) and Joreskog and Yang (1996), a Bayesian approach and an exact ML approach. The empirical performances of these approaches are assessed using simulation studies in terms of their capabilities in producing reliable parameter and standard error estimates. It is found that whilst the Bayesian and the exact ML approaches produce satisfactory results in all the settings under consideration, and are in general very reliable; the product indicator ML approaches can only produce reasonable results in simple models with large sample sizes.

Local influence analysis of structural equation models with continuous and ordinal categorical variables.

This paper proposes a method to assess the local influence of minor perturbations for a structural equation model with continuous and ordinal categorical variables. The key idea is to treat the latent variables as hypothetical missing data and then apply Cook's approach to the conditional expectation of the complete-data log-likelihood function in the corresponding EM algorithm for deriving the normal curvature and the conformal normal curvature. Building blocks for achieving the diagnostic measures are computed via observations generated by the Gibbs sampler. It is shown that the proposed methodology is relatively simple to implement, computationally efficient, and feasible for a wide variety of perturbation schemes. Two illustrative real examples are presented.