Models for estimation of production, cost, and other frontiers have evolved within two parallel traditions, the parametric and the nonparametric approach. The parametric stochastic frontier analysis (SFA) approach (e.g., Aigner et al. (1977), Meeusen and van den Broeck (1977), Olson et al. (1980)) has focused on the development of stochastic frontier models based on parametric specifications of fUnctiona forms for the frontier and the residuals involved. The nonparametric data envelopment analysis (DEA) approach (e.g., Charnes et al. (1978, 1979), Banker et al. (1984). F&-e, Grosskopf and Love11 (1985), F&e and Hunsaker ( 1986)) has focused on the development of multiple-input and multiple-output configurations. Both approaches have been extended during the past 15 years and applied in a range of empirical settings to identify technical inefficiencies.
The ten papers included in this volume were presented by invited researchers at the Advanced Research Workshop on Efficiency Measurement held at Odense University, May 22-24, 1995. Each of the papers deals with important issues in relation to the state of the art of SFA and DEA in relation to efficiency measurement. The main purpose of the workshop was to bring together researchers within these two approaches to frontier estimation for an exchange of ideas and a constructive critique of each of the two approaches.
Besides the presentation of this collection of papers the workshop also presented an interesting discussion of a number of important issues. The discussion was recorded and this postscript presents the discussion related to some of the issues raised in connection with the discussion of the papers.