A Simple Analytic Approximation Approach for Estimating the True Random Effects and True Fixed Effects Stochastic Frontier Models

38 Pages Posted: 22 Dec 2010 Last revised: 4 Jul 2011

See all articles by Wen-Jen Tsay

Wen-Jen Tsay

Academia Sinica - Institute of Economics

Peng-Hsuan Ke

Academia Sinica - Institute of Economics

Date Written: July 1, 2011

Abstract

This paper derives an analytic approximation formula for the likelihood function of the true random effects stochastic frontier model of Greene (2005) with a time span T = 2. Gaussian quadrature procedure and simulation-based method is not required for the closed-form approach. Combining the analytic formula with a pairwise likelihood estimator (PLE), we easily can estimate the random effects stochastic frontier models with T > 2. This analytic approximation approach is also applicable to the true fixed effects stochastic frontier model of Greene (2005) after the fixed effects parameters are eliminated from the use of pairwise differencing or first differencing operators. The Monte Carlo simulations confirm the promising performance of the analytic methodology under all the configurations generated from the true random effects and true fixed effects stochastic frontier models in this paper. The proposed method is applied to the World Health Organization’s (WHO) panel data on national health care systems.

Keywords: Random Effects, Panel Stochastic Frontier Model

JEL Classification: C33

Suggested Citation

Tsay, Wen-Jen and Ke, Peng-Hsuan, A Simple Analytic Approximation Approach for Estimating the True Random Effects and True Fixed Effects Stochastic Frontier Models (July 1, 2011). Available at SSRN: https://ssrn.com/abstract=1729011 or http://dx.doi.org/10.2139/ssrn.1729011

Wen-Jen Tsay (Contact Author)

Academia Sinica - Institute of Economics ( email )

128 Academia Road, Section 2
Nankang
Taipei, 11529
Taiwan

Peng-Hsuan Ke

Academia Sinica - Institute of Economics

128 Academia Road, Section 2
Nankang
Taipei, 11529
Taiwan

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
89
Abstract Views
650
Rank
516,629
PlumX Metrics