| As stochastic frontier model of technical efficiency estimates of parameters of the method, originally developed by Aigner, Lovell, Schmidt, Meeusen, Broeck, Battese, Corra three groups of researchers in 1977 also proposed. And quickly developed into an important branch of econometrics, provides research methods and analytical tools for the industry technical efficiency. After thirty years of development, in the form of a series set on stochastic frontier model parameter estimation methods and theoretical results of technical efficiency inferred emerging.With the in-depth development of economic globalization, the economy of any one region can not exist independently, it is always inextricably linked with the presence of other economies, when a disturbance caused by exogenous shocks to the economy of a region, which impact species often spread out, spread to neighboring areas and even further afield. This year, the spatial interaction and spatial structure of the main content of the rapid development of spatial econometrics and widely used in the field of regional science, environmental economics, and other infectious diseases. In the presence of spatial relationships, if still assumed sectional between independent, technical efficiency parameters and inference, and will inevitably lead to the actual situation is not consistent.The main innovation of this article is to further the theories and methods of spatial econometrics into the stochastic frontier analysis framework, due to space considerations variable lag, assuming technical inefficiency items subject to half-normal distribution. Gibbs sampling Bayesian inference has been combined with the point estimate conditional posterior distribution of the model parameters, push had technical efficiency. Monte Carlo simulations show that the estimates are very close to the true value, the Bayesian approach used in this paper on the spatial lag stochastic frontier model to infer a good effect. |
PDF Full Text Request