Title: Firm heterogeneity and dynamic panel stochastic frontier Models
Authors: Hung-pin Lai - National Chung Cheng University (Taiwan) [presenting]
Abstract: The estimation of dynamic panel stochastic frontier (DSF) models with firm heterogeneity is considered. We assume that the technical inefficiency follows an AR(1) process, which allows the firm to improve its efficiency from past experiences. Moreover, the model contains two sources of unobserved firm heterogeneity. One is from the heterogeneous distribution of the inefficiency. The other is either from the firm fixed effects, which implies heterogeneity in the production technology; or from the heterogeneity in the speed of adjustment of inefficiency. In the latter case, the AR(1) coefficient is firm specific. We discuss using the likelihood-based approaches to estimate the models. The finite sample performance of the proposed estimators is also investigated by Monte Carlo experiments.