Partial Projective Resampling Method for Dimension Reduction: With Applications to Partially Linear Models

• Activity Number: 36
• Type: Contributed
• Date/Time: Sunday, July 31, 2016 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Learning and Data Science
• Author(s): Haileab Hilafu* and Wenbo Wu
• Companies: University of Tennessee and University of Oregon
• Keywords: Partial central subspace ; Partially linear models ; Projective Resampling ; Sucient dimension reduction

Abstract

In many regression applications, the predictors naturally fall into two categories: "predictors of primary interest" and "predictors of secondary interest". It is often desirable to have a dimension reduction method that focuses on the predictors of primary interest while controlling the effect of the predictors of secondary interest. To achieve this goal, we propose a partial dimension reduction method via projective resampling of a composite vector containing the response variable and the predictors of secondary interest. The proposed method is general in the sense that the predictors of secondary interest can be quantitative, categorical or a combination of both. A special application of the proposed method for estimation in partially linear models is emphasized. The performance of the proposed method is studied through simulations, and its usefulness is demonstrated by applications to two real datasets.