Hello Statalists,

I have a balanced panel data of 131 countries over 20 years with fixed effect (using xtreg) and I found serial correlations (using xtcsd) and heteroskedasticity (using xttest3) in the model. I tried correcting the model by using
Code:
xtreg <depvar> <indepvars>, fe vce(cluster panelid)
but it turned out all of independent variables are not significant (which were significant before). So, I would like to know what should I do about this or there is any other method to fix those two problems rather than cluster? Any help would be much appreciated

Thanks