This paper presents a new stochastic frontier (SF) model for panel data. The new model moves the SF approach one step further by taking into account unobserved firm heterogeneity, short- run and long-run inefficiency. By doing so the model can not only separate firm heterogeneity from long-run (persistent) inefficiency, but it can also estimate both short-run and long-run inefficiency. Previous panel data models either confounded persistent inefficiency with firm effects (heterogeneity) or firm effects were incorrectly treated as persistent inefficiency. Both formulations are misspecified and are likely to give wrong estimates of overall inefficiency. The model presented in this paper avoids this problem by disentangling persistent inefficiency com- ponent from firm-effects while accommodating short-run inefficiency. Each of these components is treated as independent random effects. We use results from closed-skew normal distribution to derive both the log-likelihood function of the model in closed form and the posterior expected values of the random effects. These posterior expected values are used to estimate short-run and long-run (in)efficiency as well as random firm effects. The proposed model is general enough to nest all the currently used panel SF models and thus appropriateness of these models can be tested against the general model empirically. We provide empirical results from three different applications using our general model as well as several popular models that are currently used in the literature.

(2012). A stochastic frontier model with short-run and long-run inefficiency [conference presentation - intervento a convegno]. Retrieved from http://hdl.handle.net/10446/32718

A stochastic frontier model with short-run and long-run inefficiency

COLOMBI, Roberto;MARTINI, Gianmaria;
2012-01-01

Abstract

This paper presents a new stochastic frontier (SF) model for panel data. The new model moves the SF approach one step further by taking into account unobserved firm heterogeneity, short- run and long-run inefficiency. By doing so the model can not only separate firm heterogeneity from long-run (persistent) inefficiency, but it can also estimate both short-run and long-run inefficiency. Previous panel data models either confounded persistent inefficiency with firm effects (heterogeneity) or firm effects were incorrectly treated as persistent inefficiency. Both formulations are misspecified and are likely to give wrong estimates of overall inefficiency. The model presented in this paper avoids this problem by disentangling persistent inefficiency com- ponent from firm-effects while accommodating short-run inefficiency. Each of these components is treated as independent random effects. We use results from closed-skew normal distribution to derive both the log-likelihood function of the model in closed form and the posterior expected values of the random effects. These posterior expected values are used to estimate short-run and long-run (in)efficiency as well as random firm effects. The proposed model is general enough to nest all the currently used panel SF models and thus appropriateness of these models can be tested against the general model empirically. We provide empirical results from three different applications using our general model as well as several popular models that are currently used in the literature.
2012
Colombi, Roberto; Kumbhakar, Subal; Martini, Gianmaria; Vittadini, Giorgio
File allegato/i alla scheda:
File Dimensione del file Formato  
long_short_14.pdf

Solo gestori di archivio

Descrizione: author's postprint - versione referata
Dimensione del file 505.65 kB
Formato Adobe PDF
505.65 kB Adobe PDF   Visualizza/Apri
Pubblicazioni consigliate

Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10446/32718
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact