Semiparametric Regression with a Mixture Model with Both Known and Unknown Components

• Activity Number: 353 - Contributed Poster Presentations: Section on Nonparametric Statistics
• Type: Contributed
• Date/Time: Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Weibin Zhong* and Ao Yuan and Kepher H Makambi
• Companies: Georgetown University Medical Center and Georgetown University and Georgetown University
• Keywords: Empirical likelihood ; regression ; robustness ; side information

Abstract

We propose a semiparametric method of regression parameters in a mixture model, in which the true model is known only to some extent, and so the true model is a mixture with a known component and an unknown component. We use empirical likelihood method to model the unknown part, and maximum likelihood estimate for the regression parameters. Basic asymptotic properties will be studied. Simulations will be conducted to evaluate the performance of the proposed method, and the method will be applied to a real life problem.