The purpose of this sophisticated and exhaustive text is twofold: to bring together the nonparametrics literature into a single volume and to parallel traditional parametric econometrics in a “nonparametric” way.

After the obligatory introductory chapter, the authors present erudite coverage of methods of density estimation. Chapter 3 contains exhaustive coverage of conditional moment estimation. Chapter 4 details nonparametric estimation of derivatives.

Chapter 5 is a departure in focus. While the preceding chapters dealt with estimation of conditional moments of Y given that X equals some value of x and the derivatives of the measure, chapter 5 deals with models that are difficult to describe in a parametric mode. The authors refer to this scenario as “semiparametric” estimation.

Chapter 6 begins a discussion of semiparametric and nonparametric estimation of equations in more than one variable. In chapter 7, the authors approach the whole notion of nonparametrics in terms of qualitative rather than quantitative methods.

Chapter 8, “Semiparametric Estimation of Selectivity Models,” deals with a universal issue of statistics, namely randomness. In this chapter, the authors explain the processes needed for random selection.

Chapter 9 deals with the semiparametric estimation of censored regression models. Such censoring of a variable whose behavior is to be explained is not uncommon in economics problems. The chapter details parametric as well as nonparametric estimators.

Chapter 10, “Retrospect and Prospect,” is a wrap-up of the philosophy of the entire text. The authors discuss the importance of nonparametrics and stress the importance of the topic to future statistical development.

The book is not for the casual or beginning student. It will appeal to the applied statistician and the graduate student. The treatment is thorough and elegant.