> For graphing purposes I want to aggregate a panel data set, from unit
> to group level, to get a mean value for each group, each year.
> Nonetheless, I would also like to have 95 %-confidence intervals (CI)
> around the calculated mean. What is making things a bit complicated is
> that I need to use (importance) weights when aggregating the data set.
>
> One way of doing this is by using the -statsby- program together with
> the -ci- program, but this does not allows importance weights.
What about this?
*---------- begin example ---------------
sysuse auto, clear
statsby _b _se e(df_r), by(rep78) : ///
regress mpg [iw = weight]
rename _b_cons mean
rename _se_cons se
rename _eq2_stat_1 df
gen lb = mean - invttail(df,0.025)*se
gen ub = mean + invttail(df,0.025)*se
*------------- end example --------------
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
