I am looking for a longitudinal dataset with missing values that would benefit of multiple imputation techniques. Data should be relative to epidemiological studies or clinical trials and will be used in a methodological work. Do you have any idea of freely available data? Do you have a dataset that can be used? Thanks.
Hi Paolo,
you may have a look at the following webpages for longitudinal data examples:
http://tigger.uic.edu/~hedeker/long.html
http://www.biostat.jhsph.edu/~fdominic/teaching/LDA/lda.html#data
I think, however, that these datasets have full information and that your are required to artificially create missing values on your own. But there are a number of programs in the web which can do this for you.
Benjamin
Hi Paolo,
I agree with Benjamin - you can use any longitudinal data set with complete data and then randomly generate missing values on your own. You have the benefit of knowing what the true values are, so that you can determine the accuracy of the MI approach you're using.
Here is a software you may use for your study.
IVEware: Imputation and Variance Estimation Software
http://www.isr.umich.edu/src/smp/ive/
- Badges
- Science topic
- Similar topics
- Correction