Panel DOLS and Panel Causality Qns

[SporeSherry]

I am doing a study for long-run cointegration and short-run dynamics. This is followed by a causality test. All of my data used are panel data.

I have 2 questions:
1) What is the difference between selecting between the Pooled or Grouped for the Dynamic DOLS test? Both derives different coefficients, and I dont understand the implications of using them.

2) As for the Panel Causality, I am using the test on Eviews 8 that I have. According to my literature study, the Dumitrescu Hurlin test is for short-run causality. However, can I also make the assumption that in the long-run my coefficient will be the same and treat the "Stacked test" as a long-run causality test? Will someone kindly define the meaning of stacked test for me?

I have read the PDF files and done my fair share of research but those concepts are still very foreign to me. I am trying to understand them for my research paper. Please do help me out thanks! I have checked out the forum but there hasnt been a similar question.