Addressing endogeneity when estimating stochastic ray production frontiers: a Bayesian approach

Addressing endogeneity when estimating stochastic ray production frontiers: a Bayesian approach

We propose a Bayesian approach for inference in the stochastic ray production frontier (SRPF), which can model multiple-input–multiple-output production technologies even in case of zero output quantities, i.e., if some outputs are not produced by some of the firms. However, the econometric estimati...

Full description

Saved in:
Bibliographic Details
Published in:Empirical economics Vol. 62; no. 3; pp. 1345 - 1363
Main Authors: Tsionas, Mike, Izzeldin, Marwan, Henningsen, Arne, Paravalos, Evaggelos
Format: Journal Article
Language:English
Published: Berlin/Heidelberg Springer Berlin Heidelberg 01-03-2022
Springer Nature B.V
Subjects:
Bayesian analysis
Econometrics
Economic theory
Economic Theory/Quantitative Economics/Mathematical Methods
Economics
Economics and Finance
Estimates
Finance
Insurance
Management
Noise
Prices
Profit maximization
Profits
Statistics for Business
United States > US
Stochastic ray production frontier
C11
C13
Model averaging
D24
Technical inefficiency
Endogeneity
Bayesian inference
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose a Bayesian approach for inference in the stochastic ray production frontier (SRPF), which can model multiple-input–multiple-output production technologies even in case of zero output quantities, i.e., if some outputs are not produced by some of the firms. However, the econometric estimation of the SRPF—as the estimation of distance functions in general—is susceptible to endogeneity problems. To address these problems, we apply a profit-maximizing framework to derive a system of equations after incorporating technical inefficiency. As the latter enters non-trivially into the system of equations and as the Jacobian is highly complicated, we use Monte Carlo methods of inference. Using US banking data to illustrate our innovative approach, we also address the problems of missing prices and the dependence on the ordering of the outputs via model averaging.
ISSN:0377-7332
1435-8921
DOI:10.1007/s00181-021-02060-0