A Robust Iterative Algorithm for Regression Analysis of Data with Nonresponse

• Activity Number: 170
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Gong Tang*+
• Companies: University of Pittsburgh
• Keywords: missing data ; regression ; robust algorithm

Abstract

We consider regression analysis of data with missing response values. Standard statistical methods, including likelihood-based methods and weighted estimating equations, require correct specification of a model for the missing-data mechanism. Misspeci?cation of the missing-data model often causes biased estimates and wrongful conclusions. The expectation-maximization (EM) algorithm is an iterative algorithm that is often used to ?nd the maximum likelihood estimate for the likelihood-based methods. In the E-steps, given a current estimate and a model for the missing-data mechanism, the conditional expectations of the su?cient statistics are calculated. Under the premise that the current estimate is consistent, we found that those conditional expectations could be approximated from the empirical data without the need for modeling the missing-data mechanism. Subsequently we proposed a modi?ed EM algorithm regardless of the potential missing-data mechanism. The consistent initial value can be either obtained from an external complete dataset, complete recall on a subset or a robust method requiring less assumption on the missing-data model. We will present this method in the context of