面板数据用xtgcause做granger因果检验之后，如何分析？

纯粹主体 发表于 2018-4-11 15:50
图片好像打不开，见楼上

原假设为absdac不是granger因果检验，Z-bar值P值在1%情况下显著，拒绝原假设，个人理解该结果表示为absdac是granger因果检验。

找到了这篇文章：The influence of corruption on environmental sustainability in the developing economies of Southern Africa. doi: 10.1016/j.heliyon.2020.e04387

里面说：To summarise, the statistic values explained above the Granger causality process output reports the values obtained for W(W-bar), Z (Z-bar), and Ž(Z-bar tilde). Dumitrescu and Hurlin (2012) express that if N is large but T is small then Ž should be favoured. Therefore, in this paper, I test for Granger causality in the panel set by employing Dumitrescu and Hurlin (2012) xtgcause command in the Stata package using specifically the bootstrap procedure as it is also able to solve problems associated with cross-sectional dependence (Lopez and Weber, 2017).

简单说就是：如果是大N小T，用Z-bar tilde

求问格兰杰检验xtgcause不出统计结果怎么办啊？