Pitfalls of normal-gamma stochastic frontier models

Although conceptually pleasing, normal-gamma frontier models lead to difficult estimation problems. It is shown here that unless the sample size reaches several thousands of observations the shape parameter of the gamma density is hard to estimate, and that this carries over to estimates of the stochastic frontier, the individual ineciencies, and the allocation of the overall variance to the stochastic frontier and to the inefficiencies.

• metadata
• relations

Document type	Document de travail (Working Paper)
Access type	Accès libre
Publication date	1994
Language	Anglais
Number of pages	28 pages
Collection	STAT Discussion Papers - 9417, CORE Discussion Paper - 1994/41
Affiliations	UCL - EUEN/CORE - Center for operations research and econometrics UCL - EUEN/STAT - Institut de statistique
Keywords	Identiability ; least squares ; likelihood ; prole ; simulation.
Links	http://sites.uclouvain.be/IAP-Stat-Phase-V-VI/ISBApub/ISBAdp.html#1994 http://hdl.handle.net/2078.1/120775[Handle]

See also
	[boreal:46654]	Pitfalls of normal-gamma stochastic frontier models