Partial Rank Regression Models with Penalized Warped Wavelets

• Activity Number: 129 - Quantile and Nonparametric Regression Models
• Type: Contributed
• Date/Time: Monday, July 31, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Eddy Kwessi*
• Companies: Trinity University
• Keywords: Partia-Rank ; Regression ; Penalized ; Warped ; Wavelets

Abstract

In this paper, we study partial linear models with a linear part and a nonlinear monotone part. We estimate the linear part using rank regression techniques, and we estimate the nonlinear function monotone function using warped wavelets. The resulting wavelets coefficients are thresholded so as to produce a consistent estimator in nonparametric settings. Simulations as well as real data applications are provided.