Hi,
I am doing some regressions and need to correct my standard errors for any possible clustering effects (municipalities, districts).
Reading Austin's presentation (http://repec.org/usug2007/crse.pdf), it says that I can use svy commands in stata. I have tried the following and I get:
svyset goverurban || clusternum
Note: stage 1 is sampled with replacement; all further stages will be ignored
pweight: <none>
VCE: linearized
Single unit: missing
Strata 1: <one>
SU 1: goverurban
FPC 1: <zero>
Which means Stata ignores the second clustering effects (clusternum: districts)?
Has anyone had a different outcome? So it is not possible to correct for nested clusters?
thanks,
Gaby
*
* 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/
