CMStatistics 2018: Start Registration
View Submission - CMStatistics
B1763
Title: A two-stage estimation procedure for non-linear structural equation models Authors:  Klaus Holst - A.P. Moller-Maersk (Denmark) [presenting]
Esben Budtz-Joergensen - University of Copenhagen (Denmark)
Abstract: Applications of structural equation models (SEMs) are often restricted to linear associations between variables. Maximum likelihood (ML)estimation in non-linear models may be complex and require numerical integration. We introduce a simple two-stage estimation technique for estimation of non-linear associations between latent variables. Both steps are based on fitting linear SEMs: first a linear model is fitted to data on the latent predictor and terms describing the non-linear effect are predicted by empirical Bayes techniques. In the second step, the predictions are included in a linear model for the latent outcome variable. We show that this procedure is consistent and identifies its asymptotic distribution. We also illustrate how this framework easily allows the association between latent variables to be modeled using restricted cubic splines, and we develop a modified estimator which is robust to non-normality of the latent predictor. In a simulation study, we compare the proposed method to MLE and alternative two-stage estimation techniques.