蓝色 发表于 2018-8-16 17:16 
提示写的很清楚阿
requires strongly balanced data你的数据不是平衡面板,或者有缺失值
您好,在我把缺失值补充之后再用LLC检验,结果显示:Levin-Lin-Chu unit-root test for growth
---------------------------------------
Ho: Panels contain unit roots Number of panels = 26
Ha: Panels are stationary Number of periods = 27
AR parameter: Common Asymptotics: N/T -> 0
Panel means: Included
Time trend: Not included Cross-sectional means removed
ADF regressions: 2.62 lags average (chosen by AIC)
LR variance: Bartlett kernel, 9.00 lags average (chosen by LLC)
------------------------------------------------------------------------------
Statistic p-value
------------------------------------------------------------------------------
Unadjusted t -16.0168
Adjusted t* -4.6824 0.0000
------------------------------------------------------------------------------
标红的显示根据AIC信息准则选择的平均滞后阶数为2.62,偏差校正的统计量为-4.6842,显著为负,故认为面板为平稳过程。想问的是,因为变量存在滞后阶数的问题,那是不是需要在模型中引入滞后项?滞后阶数是多少就需要引入多少。是这样吗?我原模型中没有滞后项,不是动态面板。