This book examines the consequences of misspecifications in econometrics
for the interpretation and properties of likelihood-based methods of statistical
estimation and inference. Professor White first explores the underlying moti-
vation for maximum-likelihood estimation, treats the interpretation of the
maximum-likelihood estimator (MLE) for misspecified probability models
and gives the conditions under which parameters of interest can be consis-
tently estimated despite misspecification. He then investigates the limiting
distribution of the MLE under misspecification, the conditions under which
MLE efficiency is not affected despite misspecification and the consequences
of misspecification for hypothesis testing and estimating the asymptotic
covariance matrix of the parameters. The analysis concludes with an exami-
nation of methods by which misspecification problems can be empirically
investigated and offers a variety of appropriate tests.
Although the theory presented in the book is motivated by econometric
problems, its applicability is by no means restricted to economics. Subject to
defined limitations, the theory applies to any scientific context in which
statistical analysis is conducted using approximate models.