Dear all
I know there are some threads on weights and their use, but it seems
that there are not appropriate weights for what a I want to do.
I am running a regression with data for different countries and the
independent variable of interest is an indicator which was previously
estimated and hence, it has different standard errors for each
country. More specifically, using a dataset for over 100 countries I
would like to run
reg y x
but x is an indicator with different Std Errors for each observation
(country). I would like to assign more weight to those observations
(countries) with less standard error for the variable x. How should I
do? I read the Stata Manual and it seems that pweight, aweight and
fweight are not useful. The "emergency exit" would be iweights, but I
am not sure.
Thanks for your help
Diego
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
