Test for endogeneity when using in between-effects

[mona bakrrr]

Hello everyone,

I am trying to run the following model

xi: xtivreg sba ids fco vvm nan avi i.year (tnom vda = sud m_vvd m_mmd), be first

I would like to test if tnom vda are really endogenous or exogenous.

I know I can Collapse the data to a cross section and use estat endogenous but my supervisor would like me to keep the years controlled for.

I would appreciate if someone let me know the command.

Thank you very much
Mona

[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 samp[e data using dataex.

I don't see that xi: does anything (and is not needed in newer versions of Stata. When you do between, you're working on panel averages anyway so I doubt if year is doing anything. Because it is i.year, you may get the dummies, but unless you're observing different years for different panels, I don't think it should matter.

[mona bakrrr]

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 samp[e data using dataex.

I don't see that xi: does anything (and is not needed in newer versions of Stata. When you do between, you're working on panel averages anyway so I doubt if year is doing anything. Because it is i.year, you may get the dummies, but unless you're observing different years for different panels, I don't think it should matter.

Thank you very much