Hi all

I have a general question about the use of weights in regression and cluster standar error at the same time

When in run these to regressions models (same models using differents types of weights) I get different standar erros (what makes perfect sense)

Code:
reg     IHS_niha_2 sd_precipitation sd_temperature [pw=factor]
reg     IHS_niha_2 sd_precipitation sd_temperature [iw=factor]
However when I run the same regression using cluster standar error, I get the same standard error in both models.

Code:
reg     IHS_niha_2 sd_precipitation sd_temperature [pw=factor] , vce (cluster conglomera)  
reg     IHS_niha_2 sd_precipitation sd_temperature [iw=factor] , vce (cluster conglomera)
When you use cluster standard error, does STATA no longer take into account weights to calculate standard errors? If so, it makes no difference to use any type of weight (iweght vs pweight) in the regressions with cluster standar error?

My question is because a colleague who uses STATA 12 has told me that he cannot use iweghts and standard cluster error at the same time (something that i can do with STATA 15), but he can use pweghts and standard cluster error at the same time in his version of STATA

Thanks for the time