Dear All,

I was a bit confused about the validity of estimates after ivreg2. If the Kleibergen-Paap rk LM statistic and Kleibergen-Paap rk Wald F statistic indicate weak instruments but Anderson-Rubin Wald test and Stock-Wright LM S statistic reject the joint weakness of instruments, are the estimates still valid or not biased by weak instruments? Thank you for any suggestions!

Best,
Haiyan