请教各位大神,面板模型的协整检验必须三种方法必须全通过吗?我做johansen时提示错误,样本不够。我的数据时9年,辽宁省14个城市。下面是pedroni和kao检验的结果,麻烦各位前辈指点怎么看结果,哪那些值是最主要参考的。
Alternative hypothesis: common AR coefs. (within-dimension)
Weighted
Statistic Prob. Statistic Prob.
Panel v-Statistic -0.074411 0.5297 -1.785620 0.9629
Panel rho-Statistic 0.414559 0.6608 0.413875 0.6605
Panel PP-Statistic -8.310005 0.0000 -13.18354 0.0000
Panel ADF-Statistic -3.542858 0.0002 -5.937923 0.0000
Alternative hypothesis: individual AR coefs. (between-dimension)
Statistic Prob.
Group rho-Statistic 1.975501 0.9759
Group PP-Statistic -17.33935 0.0000
Group ADF-Statistic -5.553730 0.0000
这个事kao检验
t-Statistic Prob.
ADF -3.976408 0.0000
Residual variance 0.007940
HAC variance 0.005153
小弟新人,实在没有论坛币了,望各位前辈赐教!!不胜感激!!!