A0435
Title: Indirect inference of stochastic frontier models
Authors: Hung-pin Lai - National Chung Cheng University (Taiwan) [presenting]
Abstract: The standard method to estimate a stochastic frontier model is the maximum likelihood approach with the distribution assumptions of a symmetric two-sided stochastic error $v$ and a one-sided inefficiency random component $u$. When $v$ or $u$ has a nonstandard distribution, such as $v$ follows a generalized $t$ distribution or $u$ has a Chi-squared distribution, the likelihood function can be complicated. The aim is to use indirect inference to estimate the stochastic frontier models, where only least squares estimation is used. There is no need to derive the density or likelihood function, thus it is easier to handle a model with complicated distributions in practice. We examine the finite sample performance of the proposed estimator and also compare it with the standard maximum likelihood estimator as well as the maximum simulated likelihood estimator using Monte Carlo simulations. We found that our estimator performs quite well in finite samples.
