Abstract: There is a very extensive literature of statistical methods for the analysis of data with missing values. I'll provide a historical overview, including ad-hoc approaches, statistical models and conditions for ignoring the missingness mechanism, maximum likelihood methods, factored likelihood methods, the EM algorithm, Bayesian approaches, multiple imputation, and methods based on estimating equations. I'll discuss some recent developments in robust modeling and missing not at random models. I'll comment briefly on the differing perspectives of full probability modeling and more algorithmic machine learning approaches.
Virtual Workshop on Missing Data Challenges in Computation, Statistics and Applications
Statistical modeling and missing data
University of Michigan
Date & Time
September 08, 2020 | 9:15 – 10:15am