xtgcause interpretation

[Jordan Sydenham]

I have run a test for Granger non-causality in heterogeneous panels using the procedure proposed by Dumitrescu & Hurlin (Economic Modelling, 2012), which is available using the xtgcause Command. The test gives both the Z-bar and Z-bar tilde statistics, I was wondering which one of these I should report as they seem to indicate different results. All the P-values for the Z-bar statistic are 0.0000 whereas the P-values for the Z-bar tilde suggest that the null hypothesis should not be rejected. I have data from 8 time periods for 83 countries. I would appreciate some guidance on this issue. Thanks in advance.

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

Getting an answer on a user-written procedure depends on someone happening to use the procedure. It appears no one active on the list uses or can help you with this problem. Have you very carefully studied the D&H paper? If that doesn't help, you could contact the authors.

[Devashish Singh]

heyy did you get the answer to your question? i'm facing a similar issue

[Peter Ulu]

I have run a test for Granger non-causality in heterogeneous panels using the procedure proposed by Dumitrescu & Hurlin (Economic Modelling, 2012), which is available using the xtgcause Command. The test gives both the Z-bar and Z-bar tilde statistics, I was wondering which one of these I should report as they seem to indicate different results. All the P-values for the Z-bar statistic are 0.0000 whereas the P-values for the Z-bar tilde suggest that the null hypothesis should not be rejected. I have data from 8 time periods for 83 countries. I would appreciate some guidance on this issue. Thanks in advance.

the original paper of Lopez and Weber (2017) directs us on this issue:

"The testing procedure of the null hypothesis in (3) is finally based on Z and Z2. If these are larger than the standard critical values, then one should reject H0 and conclude that Granger causality exists. For large N and T panel datasets, Z-bar can be reasonably considered. For large N but relatively small T datasets, Z-tilde should be favored."

Hope this helps.