My command is
xi:xtabond2 olf l.olf psy l.psy i.t psym, gmm(olf psy,lag(2 2)) iv(i.t) r
where
olf is "the weeks out of labor market"
psy is "the depression level"
psym is rhe "missing dummy of depression level"
I want to find the causality of depressio on the weeks out of labor market.
The AR(2) test has no value, Stata result is
Arellano-Bond test for AR(2) in first differences: z = . Pr > z = .
I saw some paper said that it needs AR(2) to determine the validity of
lagged variables as the instrument. But it doesn't have it here. How
can I deal with it?
Thanks!
Su
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
