Varying-coefficient additive models for functional data

• 成果类型: Article
• 署名作者: Zhang, Xiaoke; Wang, Jane-Ling
• 署名单位: University of Delaware; University of California System; University of California Davis
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/asu053
• 发表日期: 2015
• 页码: 1532
• 关键词: Longitudinal Data; regression

摘要

Both varying-coefficient and additive models have been studied extensively in the literature as extensions to linear models. They have also been extended to deal with functional response data. However, existing extensions are still not flexible enough to reflect the functional nature of the responses. In this paper, we extend varying-coefficient and additive models to obtain a much more flexible model and propose a simple algorithm to estimate its nonparametric additive and varying-coefficient components. We establish the asymptotic properties of each component function. We demonstrate the applicability of the new model through analysis of traffic data.