SEMIPARAMETRIC ESTIMATION OF STOCHASTIC PRODUCTION FRONTIER MODELS

This article extends the linear stochastic frontier model proposed by Aigner, Lovell, and Schmidt to a semiparametric frontier model in whic h the functional form of the production frontier is unspecified and th e distributions of the composite error terms are of known form. Pseudo likelihood estimators of the parameters characterizing the two error t erms of the model are constructed based on kernel estimation of the co nditional mean function. The Monte Carlo results show that the propose d estimators perform well in finite samples. An empirical application is presented. Extensions to a partially linear frontier function and t o more flexible one-sided error distributions than the half-normal are discussed.