When I try to reproduce this (using the exact same do file), I get an
error
message on the next-to-last line:
---- begin output ----
[...]
. predict double iveps, res
. est store iv
. est change iv, scorevars(ivsc)
. suest iv
score variables for model iv no longer exist in the data
r(198);
end of do-file
r(198);
---- end output ----
Am I doing something wrong?
I don't know what might be going on here. The do-file I posted in my
prior message runs in Stata/SE 8.2 (not that SE should make any
difference), whether you use -ivreg2- or -ivreg-. Actually one
possibility might be that your -ivreg2- is out of date (the pscore
option was added at some point). Suggest you try using -ivreg- in my
do-file and see what you get.
My Stata is showing
. which estimates
/Applications/Stata/ado/updates/e/estimates.ado
*! version 1.0.9 10dec2004
. which suest
/Applications/Stata/ado/updates/s/suest.ado
*! version 2.0.8 25mar2003
