Stochastic Frontier Analysis

Modern textbook presentations of production economics typically treat producers as successful optimizers. Conventional econometric practice has generally followed this paradigm, and least squares based regression techniques have been used to estimate production, cost, profit and other functions. In such a framework deviations from maximum output, from minimum cost and cost minimizing input demands, and from maximum profit and profit maximizing output supplies and input demands, are attributed exclusively to random statistical noise. However casual empiricism and the business press both make persuasive cases for the argument that, although producers may indeed attempt to optimize, they do not always succeed. This book develops econometric techniques for the estimation of production, cost and profit frontiers, and for the estimation of the technical and economic efficiency with which producers approach these frontiers. Since these frontiers envelop rather than intersect the data, and since the authors continue to maintain the traditional econometric belief in the presence of external forces contributing to random statistical noise, the work is titled Stochastic Frontier Analysis.

'Stochastic Frontier Analysis is an excellent, up-to-date and complete summary of both the theory and techniques used in frontier production, cost and profit analysis. The authors are two of the leading researchers in the field. Students and practitioners in this area will find this book to be the only reference source they need.' William H. Greene, Stern School of Business, New York University, 'Every so often a book is written that pulls together a large subject matter in an accessible way, covering both theoretical and applied developments. This is such a book. The contribution of this volume is that this large but complex literature is presented in a manner accessible to the general applied economist. It is must reading for anyone who measures firm production, cost functions, or profit functions. Those measuring productivity change will gain new enlightenment on how to better approach this important measurement problem. Furthermore, the book could be used as a text or supplementary text in an advanced course in production, productivity change, or econometrics. It is well worth the investment and will make a valuable and useful addition to one's library.' James L. Searle, Jr, Journal of Economic Literature, 'This book is well written, comprehensive in its breadth and depth of coverage, and authoritative in its scholarship. It is certainly going to be the standard reference on stochastic frontier models for the foreseeable future.' Peter Schmidt, The Economic Journal, 'This is the best book that exists on the topic of stochastic-frontier analysis. It is clearly an excellent book that concisely reports the state of the art of this approach to efficiency analysis. This is a well-written reference book for researchers in this field and for research students considering using stochastic-frontier analysis. Libraries should carry a copy. As the first book devoted exclusively to this topic, the authors have set a very high standard.' Kevin J. Fox, Economica, From the hardback: 'Stochastic Frontier Analysis is a complete textbook on stochastic frontier models. It deals with stochastic frontier production, cost and profit functions. Models for cross-sectional and panel data are presented and discussed. The analysis of technical change and inefficiency change with panel data is also discussed in detail. The authors are highly qualified to produce such a comprehensive book on stochastic frontiers. Lovell was one of the authors of the seminal paper on the stochastic frontier production function model. He has been undertaking research on many aspects of parametric and non-parametric frontier modelling and efficiency measurement for over twenty-five years. He is a 'pioneer' in this important research area. Kumbhakar has written outstanding papers on stochastic frontier models, both of a theoretical and applied nature, over the last 12 or more years. It will surely be a helpful and widely-cited book for researchers in the field.' George E. Battese, University of New England, Australia