Offering students a unifying theoretical perspective, this innovative text emphasizes nonlinear techniques of estimation, including nonlinear least squares, nonlinear instrumental variables, maximum likelihood and the generalized method of moments, but nevertheless relies heavily on simple geometrical arguments to develop intuition. One theme of the book is the use of artificial regressions for estimation, inference, and specification testing of nonlinear models, including diagnostic tests for parameter constancy, series correlation, heteroskedasticity and other types of misspecification. Other topics include the linear simultaneous equations model, non-nested hypothesis tests, influential observations and leverage, transformations of the dependent variable, binary response models, models for time-series/cross-section data, multivariate models, seasonality, unit roots and cointegration, and Monte Carlo methods, always with an emphasis on problems that arise in applied work. Explaining throughout how estimates can be obtained and tests can be carried out, the text goes beyond a mere algebraic description to one that can be easily translated into the commands of a standard econometric software package. A comprehensive and coherent guide to the most vital topics in econometrics today, this text is indispensable for all levels of students of econometrics, economics, and statistics on regression and related topics.

Russell Davidson's profile page

James G. MacKinnon's profile page

Russell Davidson and James G. MacKinnon are both at Queen's University.

"Coverage of the geometry of least squares is excellent."--Doug Steigerwald, University of California, Santa Barbara

"An important reference source for both the theoretical and applied researcher....More importantly, the authors' view of the areas presented is cohesive, and they provide an open-ended discussion, so that the book can serve as a source of research topics as well as a reference. From this standpoint, it is very good reading for a doctoral student....Davidson and MacKinnon's book is sure to have an impact on the way econometrics is taught; my hope is that the geometric approach, widely and quite consistently used by the authors, will be adopted in the exposition of regression, illustration of the classical test statistics, and examination of test power. Certainly, the tool of projection orthogonally to part of the regression space (the Frisch-Waugh-Lovell theorem) should be adopted more widely for its convenience in simplifying many derivations."--Econometric Theory

"Well-written advanced textbook in econometrics, suitable for seminar courses. With its lucid analysis, it emerges as an extremely useful tool for applied econometricians."--Madhu Mohanty, California State University

"A wonderful text. The book is comprehensive and has a most authoritative discussion of topics of current interest such as cointegration, nonlinear simultaneous equation models, specification testing, etc."--Sunil Sapra, California State University at Los Angeles

"This is a unique and fascinating book. It's the only econometrics textbook that has ever given me the urge to read it from cover to cover."--Stratford Douglas, West Virginia University

"Clearly written and makes clear a lot of links between different estimation procedures."--Curtis J. Simon, Clemson University

"This is the most up-to-date econometrics textbook. It deals with topics which were so far discussed only in journal articles....A must book for any higher level graduate econometrics course."--Professor Anil K. Bera, University of Illinois

"Great book! Good reference for anyone wishing to get an overview of the state of the art. Good pace, topic selection, level of difficulty. Also, good use of notation."--Dean Allen Schiffman, University of California, San Diego

"Extremely valuable in the sense that it balances the coverage between test of hypothesis and estimation. Most books treat test of hypothesis as a side issue. The book is well-contained and easy to read. An excellent textbook."--Choon-Geol Moon, Rutgers University

"Good coverage of standard econometric theory."--M.M. Ali, University of Kentucky