Bayesian methods are moving into structural equation modeling. The most sophisticated approach to modeling interactions is Bayesian. People who want to be able to predict the values of observed variables need a Bayesian approach.
This book, with the code and datasets available from the publisher's website, will help you to estimate SE models using the Bayesian approach and the free WinBUGS software. Yes, it's a math-heavy book, but Sik-Yum Lee does a great job explaining this very different approach. Lee demonstrates Bayesian methods applied to basic models, interaction models, mixture models, multi-level models, and models with non-normal distributions. You really want to have this book, if you are a serious SEM user.

Читать «Structural Equation Modelling: A Bayesian Approach (Wiley Series in Probability and Statistics)»

Купить электронную версию