Using xtdcce2 for PMG

[Ronak Parikh]

Hi,

I was thinking of running a PMG regression through the xtdcce2 command - I had a question regarding this: does this mean that the output is then robust to cross-sectional dependence? If so, when writing about the model, is it then acceptable to say that I used PMG that is robust to cross-sectional dependence?

Thanks,

Ronak

[JanDitzen]

Hi Ronak,
if you use xtdcce2 to estimate a model with long run coefficients, you remove potential cross-section dependence - this is the idea behind the CCE estimator. So yes, you are right to say this. However please be aware of the difference between the CCE-ECM and the PMG estimator - please see the help file for a discussion.
Hope this helps.
Jan