Content area
Full Text
The Econometrics of Financial Markets
By John Y. Campbell. Andrew W. Lo, and A. Craig MacKinlay, Princeton, NJ, Princeton University Press, 1997, pp.320, $49.50.
The remarkable growth of interest in financial markets in the past twenty years has resulted in a large outpouring of books and articles, both theoretical and empirical, analyzing the functioning of these markets. The development of this literature has been characterized by the introduction of sophisticated statistical techniques now commonly used by finance professionals. The Econometrics of Financial Markets, justly called a "landmark" by some reviewers, is the first survey and exposition of the existing empirical literature presented in the form of a textbook. It represents an impressive intellectual achievement of three highly regarded scholars who themselves made a number of significant contributions to the financial literature. It also represents a tremendous amount of work; the authors must have read - and studied - some 830 articles, research papers and books referenced in their text.
The book is presented as a textbook intended for graduate students and industry professionals interested in financial modeling. It is a demanding book because, as the authors themselves declare, "rewards of financial econometrics come at a price." The readers are expected to have a solid background in mathematics, probability, statistics, and financial theory. The work covers virtually all topics of the empirical finance literature, with each chapter starting with an outline of the relevant theory, followed by an empirical demonstration, and ending with a set of problems and exercises.
The introductory chapter contains a brief description of the background knowledge expected of the reader, a discussion of notation, basic definitions and conventions, and an eloquent but inconclusive section on market efficiency. The second chapter deals with the question of predictability of asset returns, concentrating on the random walk hypotheses. The authors restrict the discussion...