Exact inference for errors-in-variables regressions

Abstract: When some or all regression covariates are observed with error, the standard least squares estimator is inconsistent.  Several available techniques can correct this inconsistency, but all pay a price; some require additional observations be collected, and others lack validity for small samples.  We present a new technique, using inferential models, to provide valid, probabilistic inference in linear regressions with errors in variables, with initial results showing the potential to outperform current methods.  This is joint work with Dr. Ryan Martin.

Host: Jose Figueroa-Lopez