Heteroscedasticity in Stochastic Frontier Models: a Monte Carlo Analysis.

This paper uses Monte Carlo experimentation to investigate the finite sample properties of the maximum likelihood (ML) estimators of the half-normal stochastic frontier production functions in the presence of heteroscedasticity. It is found that when heteroscedasticity exists correcting for it leads not only to a substantial improvement of the statistical properties of estimators but also to improved efficiency and ranking measures. On the other hand correcting for heteroscedasticity when there is none has serious adverse results. Hence, there is a need for testing for heteroscedasticity and if there is any the appropriate correction should be made.