A comparison of alternative parametric efficiency estimates using rank-sum test statistic

The aim of this paper is to compare the empirical results from three alternative parametric efficiency models using rank-sum test statistic. The comparison involves the technical efficiency scores and their hospitals ranking of the following models: a) Battese and Coelli (1992); b) technical inefficiency effect model (Battese and Coelli, 1995); c) non-neutral frontier model (Huang and Liu, 1994). For all models an output – distance production function are estimated using panel data of 112 Greek public hospitals. Technical efficiency scores found to be from 37.10% to 58.10%. Battese and Coelli (1992) model represented higher proportion of hospitals with technical efficiency scores between 50%-70%. Rank-sum test statistic implied that the three stochastic models came from different distribution.