格兰杰因果关系检验在面板数据上的应用研究  

Applied research of granger causality tests in panel data

作　　者：林晓羽[1] 

机构地区：[1]福州大学至诚学院,福建福州350003

出　　处：《福州大学学报（自然科学版）》2010年第2期183-187,共5页Journal of Fuzhou University(Natural Science Edition)

摘　　要：基于格兰杰因果关系检验在面板数据研究成果基础上,建立一种设定检验方法去选择格兰杰因果关系方程,并对固定效应和随机效应的选择进行描述,建立了一套比较完整的应用理论框架.同时指明模型在设定中的应该注意的数据平稳性和协整分析、滞后期的选择.Based on the research results of granger causality tests in panel data,this article is to establish a test technique for choosing the four types equations of granger causality test,namely,fixed coefficient model,random coefficient model,mixed fixed-random coefficient model and mixed randomfixed coefficient model,and furthermore,to discuss the difference between fixed-effects model and random-effects model in granger causality test and to choose one of them by Hausman test. At the same time,this paper point out a matter needing attention： the analysis of time series stationarity and contegration,the choice of lag.

关 键 词：格兰杰因果关系检验 面板数据 Hausman设定检验 

分 类 号：F224.0[经济管理—国民经济]