29 Nov
2003
29 Nov
'03
3:59 p.m.
Hi,
would anyone be able to help with a longitudinal, cross national dataset?
My full model contains 90 countries observed on 18 variables for 10
years - but missing cases reduce the model to 71 countries. Is there a
way to make multiple imputation work? Amelia goes
through all the steps but the last one, returing an error message saying
that "columns don't match." Is the dataset simply too large? I have tried
the options for TSCS data without specifying country or time dummies, but
the procedure still stops. Any help would be truly appreciated!
Simone Polillo