Applied Statistical Methods for Incomplete Data Analysis

Course Description

Inevitable existence of missing data is often a complicating factor in statistical analyses as most standard statistical methods are not designed to handle missing values. Given that almost all data have missingness, advanced statistical computation techniques and methods for analyzing incomplete data have developed tremendously since the 1980s. These methods have been increasingly applied in many settings including highly relevant survey data in public health (e.g., National Health and Nutrition Examination Survey) as well as in many other settings including clinical trials, administrative datasets (e.g., cancer registries) and observational studies. Statistical literature has repeatedly shown that if no action is taken about the missing data, all inferences can be grossly misleading. This course aims to teach and employ principled methods of incomplete data analysis to arrive at objective inferences using highly accessible software tools in SAS as well as R.

Code

EPBI 8304

Credits

3

Subject

EPBI

Programs That Include This Course

• Doctoral Doctor of Philosophy in Epidemiology
• Master's Master of Public Health: Applied Biostatistics
• Master's Master of Science in Public Health Data Science