I would like to test for cross-sectionally dependent residuals in my panelmodel (pooled). I found pcdtest() in plm package and tried pcdtest(mod, test="cd"), which yielded p = 0.03845. However, pcdtest(mod, test="lm") gave me p = 0.2266.

According to plm documentation test="cd" uses Pesaran's CD statistic, while test="lm" uses Breusch/Pagan's LM statistic.

What is the difference between test="lm" and test="cd", in practical terms?


1 Answer 1


From Baltagi (2005, 3rd edition), p. 247:

Monte Carlo experiments show that the standard Breusch–Pagan LM test performs badly for N > T panels, whereas Pesaran’s CD test performs well even for small T and large N.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.