One broad answer is to
* get (or simulate) data like those you work with
* simulate realistic fractions of missing data
using mechanisms you think are realistic
* see how good -mvis- is at reproducing what
you have missed.
Deciding "how good" depends on criteria which
you can establish for yourself, or copy from
those popular in your field.
Nick
[email protected]
Jose Valderrama
> Due to problems with missing-data in a database of cross
> sectional I am
> using the command mvis. My question is if pre-established steps exist
> using the commands contained in ST0042. My question concretely is that
> or which commands to use to evaluate the quality of the imputation.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
