Semiparametric Rank Regression with Missing Responses

• Activity Number: 252
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Huybrechts Bindele*+ and Asheber Abebe
• Companies: University of South Alabama and Auburn University
• Keywords: Wilcoxon estimato ; strong consistency ; asymptotic normality ; Missing at random ; Inverse probability ; Simple imputation

Abstract

In this paper, we consider a semi-parametric regression model with responses missing at random and we study the rank estimator of the regression coefficient. Consistency and asymptotic normality of the proposed estimator are established. Monte Carlo simulation experiments show that the proposed estimator is more efficient than the least squares estimator whenever the error distribution is heavy tailed or contaminated. When the errors follow a normal distribution, these simulation experiments show that there are scenarios under which the rank estimator is more efficient than its least squares counterpart whenever a large proportion of the responses are missing.