题目：Regularized GLS and Monte Carlo Probability for Improved Structural Equation Model Evaluation

主讲人：Professor Peter M.Bentler

摘要：

General models and methods for testing causal and latent variable hypotheses on non-experimental data have been a huge success story for the behavioral sciences, especially as path analysis, confirmatory factor analysis, and general mean/covariance structure models have become easily accessible via computer programs such as EQS, Lavaan, LISREL, Mplus, etc. However, even with 50 years of technical statistical developments, the scientific conclusions reached from such analyses can be inadequate when ideal data conditions such as multivariate normality and large sample sizes are not available. This talk provides an overview of two new model testing methodologies that seem to be a substantial improvement over existing methods in small to medium sample sizes: (1) Regularized generalized least squares for normally distributed data, and (2) Monte Carlo simulated probability values for arbitrarily distributed data. These methods, implemented in EQS 6.4, are fully described in the references below. Arruda, E. H., & Bentler, P. M. (2017). A regularized GLS for structural equation modeling. Structural Equation Modeling, 24, 657-665. Jalal, S., & Bentler, P. M. (2018). Using Monte Carlo normal distributions to evaluate structural models with nonnormal data. Structural Equation Modeling, 25, 541-557

主讲人简介：

Peter M. Bentler is an American psychologist, statistician, and distinguished professor at the University of California, Los Angeles.

In multivariate analysis and psychometrics, Professor Bentler is the developer of the structural equation modeling software EQS. His publications have over 190,000 citations as of 2017. In 2014, he was awarded the Psychometric Society Career Award. In 2015, he was elected Fellow of the American Statistical Association.