STOCHASTIC PANEL FRONTIERS - A SEMIPARAMETRIC APPROACH

This paper complements the results of Hausman and Taylor (1981) and Co rnwell, Schmidt and Sickles (1990) and generalizes Park and Simar (199 4) by examining the semiparametric efficient estimation of panel model s in which the random effects and the regressors have certain patterns of correlation. A model in which this estimator may have particular p romise is the stochastic panel frontier model. In that model inefficie ncy may be correlated with certain determinants of technology or proxi es for heterogeneity in the application of that technology. Generalize d least squares or other estimators that fail to address this dependen cy structure are inconsistent. We examine semiparametric efficient est imation for three different models based on differing dependency struc tures. Efficiency of the slope parameters and the asymptotic propertie s of the level of the frontier function are explored. We illustrate ou r new estimator in an analysis of productive efficiency between select ed North American and European airline firms after domestic deregulati on in the US and prior to recent European reforms implemented in the c ourse of EC integration. (C) 1998 Elsevier Science S.A. All rights res erved.