Multiple Imputation Using Stata is a practical guide to Stata's implementation of Multiple Imputation by Chained Equations (MICE). We'll discuss issues to consider when deciding whether to use MICE, creating imputation models that will lead to valid estimates, the process of imputing, managing imputed data, and estimation using imputed data.
To benefit from this class you'll need a solid working knowledge of Stata usage and syntax, such as that provided by our Introduction to Stata class or article series, plus familiarity with regression analysis in Stata (Stata Regression Fundamentals could get you started). You should also have a basic understanding of what multiple imputation is. The class will focus on the practical issues that arise in using multiple imputation rather than the theory behind it, though the two are, of course, linked.
Dates: 10/25, 11/1, 11/8
Time: 10:30 - 11:45
Each session of this class builds on the material taught in the previous sessions. If you cannot attend all of the class's sessions but still want to take the class, you must contact the Help Desk, find out what will be covered in the session(s) you will miss, and learn that material on your own before the next session. In most cases the material can be found in the SSCC Knowledge Base.