Gretchen Carrigan Adrian G. Barnett Annette J. Dobson Gita Mishra
Abstract
Missing data is a common problem in survey based research. There are many packages that compensate for missing data but few can easily compensate for missing longitudinal data. WinBUGS compensates for missing data using multiple imputation, and is able to incorporate longitudinal structure using random effects. We demonstrate the superiority of longitudinal imputation over cross-sectional imputation using WinBUGS. We use example data from the Australian Longitudinal Study on Women’s Health. We give a SAS macro that uses WinBUGS to analyze longitudinal models with missing covariate date, and demonstrate its use in a longitudinal study of terminal cancer patients and their carers.
Download Info
To download:
If you experience problems downloading a file, check if you have the
proper application to
view it first. Information about this may be contained
in the File-Format links below. In case of further problems read
the IDEAS help
file. Note that these files are not on the IDEAS
site. Please be patient as the files may be large.