Dear all,

I'm very new to stata and panel. I want to perform cointegration test for panel var. ex) xtwest
However, Unlike var cointegration test(ex vecrank code), xtwest or xtpedroni code have to determine "dependent variable" and "independent variables".
In this case, I would like to test all of the variables one by one by specifying them as dependent variables. Is there anything wrong with this?

ex) pvar with variable x,y,z
Code:
xtwest x y z, lags(1 2)
then result,
xtwest y z x, lags(1 2)
then result,
xtwest z x y, lags(1 2)
then result
If it's possible, how to interpret those results?

Thanks!