A hybrid method combining the PLS and the Bayesian approaches to estimate the Structural Equation Models

Abstract

The purpose of this paper is to provide a hybrid method combining the Partial Least Squares and the Bayesian approaches to estimate the Structural Equation Models. The aim advantage of this new method is to overcome the assumption of normality that is required in Bayesian approach. The results obtained from an application on simulated and on real data show that our proposed method outperforms both PLS and Bayesian approaches in terms of standard errors.