含空间自回归误差项的空间动态面板模型的检验与模拟

发布时间：2019-01-01 16:21

【摘要】：研究目标:探究含空间自回归误差项的空间动态面板数据(SDPD)模型选择和假设检验结果的水平扭曲、检验功效。研究方法:构建含空间自回归误差项的SDPD模型的空间Hausman、LM和LR检验统计量,选择蒙特卡洛模拟进行分析。研究发现:含空间自回归误差项的SDPD模型的各类检验统计量的大样本性质良好;时空滞后项对空间Hausman检验结果的影响比空间滞后项显著;条件LM_(λ|α)、LR_(λ|α)检验是含空间自回归误差项随机效应SDPD模型的最优检验统计量,时空滞后项对LM、LR检验结果的影响比自回归误差项更显著。研究创新:解析了SDPD模型中各空间关联项系数的波动对模型检验精度的影响。研究价值:探究含空间自回归误差项SDPD模型的选择问题,为实证提供理论依据。
[Abstract]:Objective: to explore the horizontal distortion and efficiency of (SDPD) model selection and hypothesis test of spatial dynamic panel data with spatial autoregressive error term. Methods: the spatial Hausman,LM and LR test statistics of SDPD model with spatial autoregressive error term were constructed and Monte Carlo simulation was used to analyze them. It is found that the large sample properties of all kinds of test statistics in SDPD model with spatial autoregressive error term are good, and the influence of space-time lag term on spatial Hausman test results is more significant than that on spatial lag term. Conditional LM_ (位-伪) and LR_ (位-伪) test are the optimal test statistics for SDPD model with random effect of spatial autoregressive error term. The influence of space-time lag term on the result of LM,LR test is more significant than that of autoregressive error term. Research innovation: this paper analyzes the influence of the fluctuation of spatial correlation coefficient on the accuracy of SDPD model. Research value: to explore the choice of SDPD model with spatial autoregressive error term to provide theoretical basis for empirical research.
【作者单位】：苏州科技大学商学院;江西财经大学统计学院;
【基金】：国家自然科学基金项目“环境约束下区域技术创新效率的空间特征溢出效应及影响机制研究”(41461025)、“环境约束下要素集聚促进区域产业结构升级的机制研究——基于区域创新能力差异化视角”(71273122)和“基于知识溢出的区域技术创新驱动与产业结构优化升级耦联机制研究”(71473109)的资助
【分类号】：O212

【相似文献】