首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Parametric approximations of nonparametric frontiers
Institution:1. Prime Minister’s Office, Department for economic policy, Via della Mercede 9, Rome I-00187, Italy;2. CNRS-LEM (UMR 9221), IESEG School of Management, 3 rue de la Digue, Lille F-59000, France;3. Research Unit MEES, KU Leuven, Warmoesberg 26, Brussel B-1000, Belgium
Abstract:The maximal achievable level of output for a given level of inputs defines the production frontier that can serve as benchmark to evaluate individual firm efficiencies. Nonparametric envelopment estimators (free disposal hull, data envelopment analysis) have been mostly used because they rely on very few assumptions, whereas parametric forms for the frontier allow for richer economic interpretation. Most of the parametric approaches rely on standard regression fitting the shape of the center of the cloud of points. In this paper, we investigated a new approach, which captures the shape of the cloud points near its boundary. It offers parametric approximations of nonparametric frontiers. We provide the statistical theory (asymptotic). Some simulated examples show the advantages of our method compared with the usual regression-type estimators.
Keywords:
本文献已被 ScienceDirect 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号