Observed and unobserved heterogeneity in stochastic frontier models

Abstract

<div lang="en" class="abs"><p class="abs_header">Abstract</p><p>Stochastic frontier modeling has proceeded rapidly recently. Heterogeneity modeling internalized into frontier estimation has opened up new promising possibilities. In this paper we study different ways of considering heterogeneity in stochastic frontier models. It is possible to take heterogeneity into account by including those effects in the mean and/or variance of the distribution of inefficiency (observed heterogeneity) or by randomizing some parameters of the stochastic frontier model (unobserved heterogeneity). We compare the advantages of heterogeneity including models over the conventional random effects models for measuring the cost efficiency of electricity distribution utilities. Our results indicate that in all heterogeneity accounting models mean inefficiency decreases significantly compared to the basic random effects model. According to our results randomizing some of the parameters seems to help to capture the unobserved heterogeneity and hence this kind of firm specific heterogeneity does not appear as inefficiency in our estimation results. Notable is that the model which accounts observed heterogeneity and the models which account unobserved heterogeneity produce clearly different rank orders.</p></div>