On double-index dimension reduction for partially functional data. Issue 3 (3rd July 2019)

• Record Type: Journal Article
• Authors: Yang, Guangren; Lin, Hongmei; Lian, Heng
• Is Part Of: Journal of nonparametric statistics. Volume 31:Issue 3(2019)
• Page Start: 761
• Page End: 768
• Publication Date: 2019-07-03
• Subjects: Functional data analysis -- regularisation -- sliced inverse regression; 62G08; Nonparametric statistics -- Periodicals; 519.5
• DOI: 10.1080/10485252.2019.1632308
• Languages: English
• ISSNs: 1048-5252

Abstract

ABSTRACT: In this note, we consider the situation where we have a functional predictor as well as some more traditional scalar predictors, which we call the partially functional problem. We propose a semiparametric model based on sufficient dimension reduction, and thus our main interest is in dimension reduction although prediction can be carried out at a second stage. We establish root- n consistency of the linear part of the estimator. Some Monte Carlo studies are carried out as proof of concept.