In real data analysis with structural equation modeling, data are unlikely to be exactly normally distributed.If we ignore the non-normality reality, the parameter estimates, standard error estimates, and model fit statistics from normal theory based methods such as maximum likelihood (ML) and normal theory based generalized least squares estimation (GLS) are unreliable. On the other hand, the asymptotically distribution free (ADF) estimator does not rely on any distribution assumption but cannot demonstrate its efficiency advantage with small and modest sample sizes. The methods which adopt misspecified loss functions including ridge GLS (RGLS) can provide better estimates and inferences than the normal theory based methods and the ADF estimator in some cases. We propose a distributionally-weighted least squares (DLS) estimator, and expect that it can perform better than the existing generalized least squares, because it combines normal theory based and ADF based generalized least squares estimation. Computer simulation results suggest that model-implied covariance based DLS (DLS M ) provided relatively accurate and efficient estimates in terms of RMSE. In addition, the empirical standard errors, the relative biases of standard error estimates, and the Type I error rates of the Jiang-Yuan rank adjusted model fit test statistic (T JY ) in DLS M were competitive with the classical methods including ML, GLS, and RGLS. The performance of DLS M depends on its tuning parameter a. We illustrate how to implement DLS M and select the optimal a by a bootstrap procedure in a real data example. ModelingStructural equation modeling (SEM) is widely used in social and behavioral research, but its statistical methodology remains marginally capable of dealing with empirical data encountered in many psychological and behavioral studies. First, statistics in SEM rely on large sample size approximation.
Distributionally-Weighted Least Squares in Structural Equation Modeling 3
Distributionally-Weighted Least Squares in Structural EquationThat is, their use relies on asymptotic properties as sample size N becomes extremely large (N → ∞).However in real data analysis, sample sizes are usually moderate or even small. Although SEM methods usually provide consistent parameter estimates and consistent standard error (SE) estimates, the estimates are not necessarily unbiased with finite sample size. Second, although data are typically nonnormally distributed (e.g., Cain et al., 2017), the mainstream estimators for SEM are still based on normal theory, such as maximum likelihood (ML) and normal theory based generalized least squares estimation (GLS).