机构地区: 天津大学管理与经济学部
出 处: 《商业经济与管理》 2009年第11期65-71,共7页
摘 要: 文章利用Monte Carlo模拟方法研究了带结构突变的面板数据单位根伪检验问题。研究发现,只有在突变前后样本数相差较大,或者均值突变不明显时,均值突变才不会导致传统面板单位根检验结果失效;趋势突变在绝大多数情况下将导致传统的面板数据单位根检验失效。 Spurious unit root tests for panel data with structural change are made by using Monte Carlo methods in this paper. The result shows that unit root tests for panel data with expectation shift are effective only when the number of samples varies greatly before and after the turning point or when expectation shift is not obvious; and unit root tests for panel data with variance shift are ineffective and will lead to spurious result in most cases.