I am working with survey data that would benefit greatly from multiple imputations. I
have used the Amelia software to generate 5 imputed datasets. However, I am having
difficulty availing myself of the benefits of the Clarify program. I would be very
grateful if you could answer some questions about the process and diagnose an error
message that I am receiving.
Question #1. I began by uploading my original dataset into STATA. I was able to run my
logit models, but the listwise deletion was a problem. I then tried to see how things
looked with the multiple imputations. Can I start with my original dataset with the
missing values, or must I begin with one of the imputed datasets? Although I could run a
logit model with my original dataset, I was unable to run one using one of the imputed
datasets.
Question #2. After uploading the five imputed datasets into STATA memory, I typed the
following into the command line:
estsimp logit influ gender age edu ideology deepsouth mayrad distelec ncounc blkperc
quality distblk distmon, mi (B:/amelia1.dta B:/amelia2.dta B:/amelia3.dta B:/amelia4.dta
B:/amelia5.dta)
STATA responded as follows:
Estimation number : 1 of 5
Dataset being used: B:/amelia1.dta
(Written by R. )
r(2000);
This error message is explained by STATA in the following way:
Return code 2000
no observations;
You have requested some statistical calculation and there are
no observations on which to perform it. Perhaps you specified
if exp or in range and inadvertently filtered all the data.
Thanks in advance for helping me address these issues. I am very grateful for your time
and your assistance. I can be reached at g-neddenriep(a)neiu.edu
<mailto:g-neddenriep@neiu.edu> .
GN
Gregory Neddenriep, J.D., Ph.D.
Assistant Professor of Political Science
Northeastern Illinois University
Show replies by date