A seminar titled " An Alternative Approach To Handling Missing Data By Deleting Completely Observed Records" is held in KAIST.
Title : An Alternative Approach To Handling Missing Data By Deleting Completely Observed Records
Date : November 22 (Thursday), 2012 14:00
Location : Wooribyul Seminar Room(#2201), Department of Electrical Engineering
Speaker :Professor Myunghee Paik/Department of Statistics/Seoul National University
Abstract:
In the literature on approaches to missing data, the prevailing paradigms are the likelihood method, imputation, and inverse probability weighting methods. We introduce a recently proposed alternative approach that is easy to implement, namely, deletion.
We demonstrate how this approach can be applied in longitudinal settings such as GEE and the Cox model with missing covariates. The method is compared to other approaches by simulation studies and data analysis from the Northern Manhattan Stroke study.
Bio:
ACADEMIC TRAINING
Seoul National University 1981 B.S. Nutrition
University of Pittsburgh 1984 M.S. Epidemiology
University of Pittsburgh 1987 Ph.D. Biostatistics