A Semiparametric Approach for Structural Equation Modeling with Ordinal Data

There is currently a lack of methods for non-linear structural equation modeling (NSEM) non-parametric relationships between latent variables when data are ordinal. To this end, semiparametric approach flexible NSEMs without parametric forms developed ordinal data. An indirect application finite mixture models (SEMM) employed the conditional expected mean endogenous variables. In context, classes not to be interpreted as groups observations belonging those classes, rather they serve means model functions locally linear which together approximate globally function. The proposed method based on hybrid direct maximization and expectation-maximization algorithms. Two simulation studies performed show that parameter estimates associated with low bias functional form satisfactorily estimated using approach.