Title: Spatial autoregressive panel stochastic frontier models with fixed effects
Authors: Kien C Tran - University of Lethbridge (Canada) [presenting]
Levent Kutlu - Georgia Institute of Technology (United States)
Mike Tsionas - Lancaster University (United Kingdom)
Abstract: The estimation of spatial autoregressive (SAR) panel stochastic frontier model with fixed effects is considered. We apply the closed skew normal results to obtain the distribution of the first-difference transformation of the composed-error and the log-likelihood function. Indirect, direct and total marginal effects and inefficiency are calculated. Extension of the model to allow for endogenous regressors is also provided. Monte Carlo simulations indicate good finite-sample performance of the proposed approach. An empirical application of the proposed model and approach is presented.