Estimation of Stochastic Frontier Cost Function Using Data Envelopment Analysis: an Application to the AT&T Divestiture
作者:
SueyoshiToshiyuki,
期刊:
Journal of the Operational Research Society
(Taylor Available online 1991)
卷期:
Volume 42,
issue 6
页码: 463-477
ISSN:0160-5682
年代: 1991
DOI:10.1057/jors.1991.95
出版商: Taylor&Francis
关键词: data envelopment analysis;goal programming/constrained regression;stochastic frontier cost function;natural monopoly;parametric and nonparametric approaches
数据来源: Taylor
摘要:
AbstractThis study develops a new use of data envelopment analysis for estimating a stochastic frontier cost function that is assumed to have two different error components: a one-sided disturbance (representing technical and allocative inefficiencies) and a two-sided disturbance (representing an observational error). The two error components are handled by data envelopment analysis in combination with goal programming/constrained regression. The approach proposed in this study can avoid several statistical assumptions used in conventional methods for estimating a stochastic frontier function. As an important application, this study uses the estimation technique to obtain an AT&T stochastic frontier cost function. As a result, this study measures technical and allocative efficiencies of AT&T production process and review its natural monopoly issue. The estimated stochastic frontier cost function is also compared with the other cost function models used for previous studies concerning the divestiture of the telephone industry.
点击下载:
PDF (6503KB)
返 回