Performance of Ordinary Least Squares and Heteroscedastic Consistent Covariance Matrix Estimators in Heteroscedastic Analysis of Covariance Models

• In Event: Roundtable Session 31
  In Roundtable Session: Alternatives to Parametric Analysis in the Contexts of Regression and Analysis of Variance

Abstract

Recently, there has been a focus on the performance of Heteroscedastic Consistent Covariance Matrix (HCCM) estimators in complex regression models. To date, the performance of HCCM estimators in traditional Analysis of Covariance (ANCOVA) designs has not been directly addressed. We will simulate several heteroscedastic scenarios for a traditional ANCOVA model with orthogonal non-orthogonal fixed covariates. Preliminary simulations indicate that under complete and partial null models, heteroscedasticity due to an orthogonal covariate alone will not affect the Type 1 error rate for the OLS test of adjusted group mean differences; however, heteroscedasticity due to both the covariate and group, may attenuate, exacerbate, or reverse the known effects of heteroscedasticity in unbalanced models.

Author

• T. Mark Beasley, The University of Alabama - Birmingham