A Monte Carlo Study of Ranked Efficiency Estimates from Frontier Models

Journal of Productivity Analysis, October 2012, Volume 38, Issue 2, pp 155-165

Posted: 15 Mar 2015

See all articles by Seth Richards-Shubik

Seth Richards-Shubik

Lehigh University - Department of Economics; National Bureau of Economic Research (NBER)

William C. Horrace

Syracuse University - Department of Economics

Date Written: October 1, 2012

Abstract

Parametric stochastic frontier models yield firm-level conditional distributions of inefficiency that are truncated normal. Given these distributions, how should one assess and rank firm-level efficiency? This study compares the techniques of estimating (a) the conditional mean of inefficiency and (b) probabilities that firms are most or least efficient. Monte Carlo experiments suggest that the efficiency probabilities are easier to estimate (less noisy) in terms of mean absolute percent error when inefficiency has large variation across firms. Along the way we tackle some interesting problems associated with simulating and assessing estimator performance in the stochastic frontier model.

Keywords: Truncated normal, Stochastic frontier, Efficiency, Multivariate probabilities

Suggested Citation

Richards-Shubik, Seth and Horrace, William C., A Monte Carlo Study of Ranked Efficiency Estimates from Frontier Models (October 1, 2012). Journal of Productivity Analysis, October 2012, Volume 38, Issue 2, pp 155-165, Available at SSRN: https://ssrn.com/abstract=2577958

Seth Richards-Shubik (Contact Author)

Lehigh University - Department of Economics ( email )

620 Taylor Street
Bethlehem, PA 18015
United States

HOME PAGE: http://www.lehigh.edu/~ser315

National Bureau of Economic Research (NBER) ( email )

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

William C. Horrace

Syracuse University - Department of Economics ( email )

Syracuse, NY 13244-1020
United States
315-443-9061 (Phone)
315-443-1081 (Fax)

HOME PAGE: http://faculty.maxwell.syr.edu/whorrace

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
231
PlumX Metrics