Adjusted Empirical Likelihood for Spatial Autoregressive Models with Spatial Autoregressive Disturbances
The adjustment empirical likelihood inference of spatial autoregressive models with spatial autoregressive errors is studied in this paper.Using adjusted empirical likelihood method,an adjusted empirical likelihood ratio statistic is constructed for spatial autoregressive models with spatial autoregressive errors.It is shown that the limit distribution of the adjusted empirical likelihood statistic is a chi-squared distribution.The advantages and disadvantages of the adjusted empirical likelihood method are compared with the general empirical likelihood method through simulations.Simulation results show that the adjusted empirical likelihood region has higher coverage accuracy and is faster in implementation than that of the general empirical likelihood.When the sample size is large,both general empirical likelihood method and adjusted empirical likelihood method can be used whether the error term follows a normal distribution or not.But when the sample size is small,it is recommended to use the adjusted empirical likelihood method.
spatial autoregressive errorspatial ARAR modeladjusted empirical likelihoodconfidence region
国家科技期刊平台
NSTL
万方数据
