Title: Functional partial linear quantile regression based on reproducing kernel Hilbert space
Authors: Peng Liu - University of Kent (United Kingdom) [presenting]
Nan Zhang - Fudan University (China)
Bei Jiang - University of Alberta (Canada)
Linglong Kong - University of Alberta (Canada)
Jianhua Huang - Texas A and M University (United States)
Abstract: Functional and nonfunctional data are often encountered simultaneously in modern experiments for example the clinical trial as well in economics. However, it is difficult to consider both data at the same time. We consider functional partial linear quantile regression, where both infinite dimension functional as well as finite dimension slope parameters are included. We study the theoretical properties under a reproducing kernel Hilbert space framework which was being proved to be very flexible and powerful. Under this framework, we also developed an ADMM algorithm which is very easy to implement in practical applications. Simulation studies and real data studies are performed to validate our propose methodology and practical applications respectively.