CDFP 2333

An Econometrician amongst Statisticians: T. W. Anderson


Publication Date: June 2022


T. W. Anderson did pathbreaking work in econometrics during his remarkable career as an eminent statistician. His primary contributions to econometrics are reviewed here, including his early research on estimation and inference in simultaneous equations models
and reduced rank regression. Some of his later works that connect in important ways to econometrics are also briefly covered, including limit theory in explosive autoregression, asymptotic expansions, and exact distribution theory for econometric estimators. The research is considered in the light of its influence on subsequent and ongoing developments in econometrics, notably confidence interval construction under weak instruments and inference in mildly explosive regressions.

Keywords: Asymptotic expansions, Confidence interval construction, Explosive autoregression, LIML, Reduced rank regression, Simultaneous equation models, Weak identification

JEL Classification Codes: C23

PDF icon d2333.pdf