I am running a fixed effects model and included time effects.
For the cross sectional dependence test I am using the Pesaran CD test.
However, I am not sure if I have to perform the test for cross-sectional dependence on the model with time effects or without time effects.
pcdtest(fixed.effects, test=c("cd") <- the test without time effects (p-value = 4.332e-14)
pcdtest(time.fixed.effects, test=c("cd") <- the test with time effects (p-value = 0.2674)