Endogenous Inputs and Environmental Variables in Battese and Coelli's (1995) Stochastic Frontier Model

This study provides a guideline for dealing with endogenous inputs and environmental variables in Battese and Coelli's (1995) stochastic frontier model, which has been used in a vast number of empirical studies. The presence of endogenous variables in Battese and Coelli's model leads to inconsistent parameter estimates; also, estimates in the studies that use their model may be biased. Our model uses limited information maximum likelihood methods to correct these biases. The Monte Carlo simulations provided in this article show that our estimators perform well