This online course, covers the modeling and analysis of incomplete multivariate or longitudinal data – data with records for which some, but not all observations are missing. Many analysis methods cannot handle the inclusion of such records, but omitting these records discards valuable information. There are a range of techniques to handle this situation, and this course goes beyond the methods covered in the course “Missing Data.” In this course, you will learn about treatments that apply when data are missing, but not at random. This is a very common situation, and when it occurs, the classic models do not apply. This course describes how “Missing at Random” counterpart models may be identified and assessed for their suitability for the “Missing Not at Random” situation.
- Week 1: Modeling Incomplete Data
- Week 2: Inverse Probability Weighting and Multiple Imputation
- Week 3: Initial Topics in Methods and Sensitivity Analysis for Incomplete Data
- Week 4: Further Topics in Methods and Sensitivity Analysis for Incomplete Data
About 15 hours per week, at times of your choosing.
INR 37,740 (assuming $ = INR 60)
Part Time/Full Time:
Statistical analysts and consultants who develop and apply statistical models that must be used in situations where data are incomplete, and the simpler models may not be applicable.
- Geert Molenberghs