New Semiparametric Regression Method with Applications to Selection-Bias Sampling and Missing Data Problems

• Activity Number: 26
• Type: Contributed
• Date/Time: Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
• Sponsor: IMS
• Author(s): Guoqing Diao*+ and Jing Qin
• Companies: George Mason University and NIH
• Keywords: Density ratio model ; Generalized linear model ; Missing covariate ; Nonparametric regression ; Selection-bias sampling ; Semiparametric regression

Abstract

We propose a new method to estimate a regression function based on the semiparametric density ratio model, which can be viewed as a generalized linear model with a canonical link function and an unspecified baseline distribution function. Under the density ratio model, the distribution of the observed data retains the same structure in the presence of selection-bias sampling or when the predictors are missing at random. Particularly, in the latter case, the new method utilizes all the available information and does not need to specify the distribution of the predictors or the missing probability. We establish large sample properties of the proposed regression estimators. Simulation studies demonstrate that the proposed estimators perform well in practical situations. An application to a real example is provided.