2019
DOI: 10.1016/j.jeconom.2019.01.014
|View full text |Cite
|
Sign up to set email alerts
|

A time-varying true individual effects model with endogenous regressors

Abstract: We propose a fairly general individual effects stochastic frontier model, which allows both heterogeneity and inefficiency to change over time. Moreover, our model handles the endogeneity problems if either at least one of the regressors or one-sided error term is correlated with the two-sided error term. Our Monte Carlo experiments show that our estimator performs well. We employed our methodology to the US banking data and found a negative relationship between return on revenue and cost efficiency. Estimator… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

1
23
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
6
2

Relationship

2
6

Authors

Journals

citations
Cited by 60 publications
(24 citation statements)
references
References 56 publications
1
23
0
Order By: Relevance
“…The model has the following form: lncit=f(),xitalicitβ+vit+uit, uit=G()t·ui;G()t=italicexp[]η()tTi, ui~N+(),μiσu,i20.25em=ui~N+()ψ0+z´i·ψ,exp()ωu0+z´u,i·ωu0.25em, vit~scriptN(),0σv,italicit20.25em=scriptN(),0exp()ωv0+z´v,italicit·ωv0.25em. We are aware that our methodology does not allow us to control the endogeneity previously mentioned that could exist between the turnout and the institutional efficiency, so that the estimate of the possible effect can only reflect a non‐causal relationship. Although Amsler, Prokhorov, and Schmidt () and Kutlu, Tran, and Tsionas () have developed econometric procedures that allow environmental variables to be endogenous in the sense that they are correlated with the specific inefficiency component, neither of these studies solved this problem in Battese and Coelli's () stochastic frontier model in a single step, which is the basis upon which we developed all the theoretical considerations that we have assumed.…”
Section: Methodsmentioning
confidence: 99%
“…The model has the following form: lncit=f(),xitalicitβ+vit+uit, uit=G()t·ui;G()t=italicexp[]η()tTi, ui~N+(),μiσu,i20.25em=ui~N+()ψ0+z´i·ψ,exp()ωu0+z´u,i·ωu0.25em, vit~scriptN(),0σv,italicit20.25em=scriptN(),0exp()ωv0+z´v,italicit·ωv0.25em. We are aware that our methodology does not allow us to control the endogeneity previously mentioned that could exist between the turnout and the institutional efficiency, so that the estimate of the possible effect can only reflect a non‐causal relationship. Although Amsler, Prokhorov, and Schmidt () and Kutlu, Tran, and Tsionas () have developed econometric procedures that allow environmental variables to be endogenous in the sense that they are correlated with the specific inefficiency component, neither of these studies solved this problem in Battese and Coelli's () stochastic frontier model in a single step, which is the basis upon which we developed all the theoretical considerations that we have assumed.…”
Section: Methodsmentioning
confidence: 99%
“…The violation of this assumption may bias efficiency estimates. A generalization of true fixed effects model ofGreene (2005aGreene ( , 2005b to the endogeneity case as done byKutlu et al (2017) may overcome such difficulties in the panel data context.…”
mentioning
confidence: 99%
“…Amsler et al (2016Amsler et al ( , 2017 relax the conditional independence assumption for " s it andũ it by using a copula approach. Kutlu et al (2018) show by simulations that if the …rm-speci…c individual e¤ects are included in the model, even if " s it andũ it are correlated conditionally, the estimates are still reasonable. When it is di¢ cult to …nd instruments for endogenous variables, one may use the copula approach proposed by Tran and Tsionas (2015).…”
Section: Ptmentioning
confidence: 94%
“…In this setting, v it and u it terms are assumed to be independent, which can be a questionable assumption in a variety of settings. Similar to Kutlu (2010), Karakaplan and Kutlu (2017a, b), and Kutlu et al (2018), we use a limited information maximum likelihood based approach to handle the endogeneity issue that occurs when the two-sided error term is correlated with the regressors or u it . The approach solves the endogeneity issue by including a bias correction term in the model.…”
Section: Ptmentioning
confidence: 99%
See 1 more Smart Citation