Alternative Approaches to Technical Efficiency Estimation in the Stochastic Frontier Model

Abstract

Estimating the stochastic frontier model and calculating technical efficiency of decision making units are of great importance in applied production economic works. This paper estimates technical efficiency from the stochastic frontier model using Jondrow, and Battese and Coelli approaches. In order to compare alternative methods, simulated data with sample sizes of 60 and 200 are generated from stochastic frontier model commonly applied to agricultural firms. Simulated data is employed to compare the alternative methods. Empirical results show a strong correlation between the alternative methods regardless of the differences in the actual values of the efficiency estimates. Mean technical efficiency is sensitive to the choice of estimation method. Analysis of variance and Tukey’s test suggest difference in means between the efficiency scores from different methods. Battese and Coelli’s approach produces more homogenous estimates of technical efficiency when compared with the Jondrow’s mean or mode approach. Our results suggest that differences in conclusion are possible when the alternative methods of measuring technical efficiency are applied.