Persistent and transient inefficiency in a spatial autoregressive panel stochastic frontier model

被引:3
|
作者
Lai, Hung-pin [1 ,2 ]
Tran, Kien C. [3 ]
机构
[1] Natl Chung Cheng Univ, Dept Econ, Taipei, Taiwan
[2] Acad Sinica, Res Ctr Humanities & Social Sci, Taipei, Taiwan
[3] Uniin Finite Samples, Dept Econ, 4401 Univ Dr W, Lethbridge, AB T1K 3M4, Canada
关键词
Maximum simulated likelihood; stochastic frontier; spatial autoregressive; persistent inefficiency; transient inefficiency; EFFICIENCY; HETEROGENEITY; DETERMINANTS;
D O I
10.1007/s11123-022-00638-z
中图分类号
F [经济];
学科分类号
02 ;
摘要
In this paper, we extend the four-component stochastic frontier model to allow for global spatial dependence via the endogenous spatial autoregressive variable. Our proposed model is more general than the model considered by (Glass et al., 2016) in the sense that we include a random effect as well as a permanent efficiency component. With the spatial autoregressive specification, our model is able to capture the asymmetric efficiency spillovers and also decompose the persistent/transient inefficiencies into direct and indirect efficiencies. Moreover, we also investigate the marginal effects of the exogenous variables on the persistent/transient efficiency. We suggest a maximum simulated likelihood method to estimate the frontier parameters of the model, and we predict the efficiencies using the simulated estimator. Monte Carlo simulations reveal that the suggested estimator performs well in finite samples. An empirical application is considered to illustrate the usefulness of our proposed model and method.
引用
收藏
页码:1 / 13
页数:13
相关论文
共 50 条