Dear Statalist,
I am estimating the following FE model
CBRit = a*X + b*(STATE by YEAR fixed effects) + (COUNTY fixed effects), cluster(COUNTY)
where CBR = county level birth rate for county i in year t.
There are 600 counties (from 12 states) and 8 years available. I am using the command xtivreg2 which supports simple fixed effects estimation with no endogeneous variables by omitting (varlist2=varlist_iv).
When I use cluster and include state by year fixed effects, stata tells
"Error: estimated covariance matrix of moment conditions not of full rank;
overidentification statistic not reported, and standard errors and
model tests should be interpreted with caution.Possible causes:
singleton dummy variable (dummy with one 1 and N-1 0s or vice versa)
fwl option may address problem."
The same error does not occurr if instead of using xtivreg2 I use reg by adding a dummy for each county (together with state by year FE)+ cluster, or if I use xtivreg2 without state by year FE.
Tank you for any suggestion
sara
*
* 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/
