Diagnostic Testing in Missing Data Models
In this paper we consider a class of missing data models which has attracted a great deal of attention in labor economics and in the analysis of various social experiments. Our interest focuses on the testing of the assumptions of homoscedasticity and lognormality in these models. The Lagrange multiplier tests proposed here are asymptotically locally most powerful with respect to small perturbations from the parametric specification of the null hypotheses, and we conjecture that they will also be reasonably powerful with respect to other deviation from the model's assumptions.
- A general approach to Lagrange multiplier model diagnostics
- Testing for parametric component of partially linear models with missing covariates
- Testing heteroscedasticity in partially linear models with missing covariates
- Score Test for Missing at Random or Not under Logistic Missingness Models
- Diagnostic tests as residual analysis
This page was built for publication: Diagnostic Testing in Missing Data Models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3339996)