Hello stata-users:
I'm having problems doing imputations in a large dataset (over 30,000 observations) using Stata/SE 10.1, here is the code:
ice study_id log_age sex edu log_weight log_height wc log_hip log_sbp
log_dbp hyp hyp_dx hyp_tx log_glucose dm dm_dx dm_tx log_tot_c log_hdl_c
log_triglycerides smk fast cvdhis [pweight=sweight], saving(temp3, replace)
boot passive(dm:log_glucose>=4.8362819\dm_tx:dm_tx==1 &
dm==1\dm_dx:dm_dx==1 & dm==1\ /*
*/ hyp:log_sbp>=4.9416424 | log_dbp>=4.4998097\hyp_tx:hyp_tx==1 &
hyp==1\hyp_dx:hyp_dx==1 & hyp==1) /*
*/ cmd(log_age log_weight log_height wc log_glucose log_hip log_sbp
log_dbp log_glucose log_tot_c log_hdl_c log_triglycerides: regress, smk
/*hyp_tx dm_tx*/ cvdh: logit, edu: ologit) /*
*/ genmiss(m) cycles(10) m(20) replace seed(515200) trace(fc3)
As you can see I'm trying to do 20 imputations but Stata stops after a few imputations (6, 7 or so) and produces the following message:
"Warning: while I was executing the following command:
uvis regress wc study_id log_age sex edu log_weight log_height log_hip
log_sbp log_dbp hyp hyp_dx hyp_tx log_glucose dm dm_dx dm_tx log_tot_c
log_hdl_c log_triglycerides smk fast cvdhis [pweight= sweight] if __000004,
///
( invalid name
r(198);"
I don't think this is a problem with the syntax or a type, because Stata fits the model several times and then stops. Moreover, I've used the exact same program, but a different seed to impute 20 copies (m=20) without problems. I hope someone could shed some light on what's going on.
Thanks,
VICTOR M. HERRERA MD. MS.
Research Assistant - Wisconsin Diabetes Registry
Population Health Sciences Department
University of Wisconsin
610 Walnut St. 626 WARF
Madison, WI 53726
(608) 265-3686
*
* 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/
