Oracally Efficient Two-Step Estimation of Generalized Additive Model

• 成果类型: Article
• 署名作者: Liu, Rong; Yang, Lijian; Haerdle, Wolfgang K.
• 署名单位: Soochow University - China; University System of Ohio; University of Toledo; Michigan State University; Humboldt University of Berlin
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459
• DOI: 10.1080/01621459.2013.763726
• 发表日期: 2013
• 页码: 619-631
• 关键词: coefficient model; regression-models; marginal integration; polynomial spline; identification

摘要

The generalized additive model (GAM) is a multivariate nonparametric regression tool for non-Gaussian responses including binary and count data. We propose a spline-backfitted kernel (SBK) estimator for the component functions and the constant, which are oracally efficient under weak dependence. The SBK technique is both computationally expedient and theoretically reliable, thus usable for analyzing high-dimensional time series. Inference can be made on component functions based on asymptotic normality. Simulation evidence strongly corroborates the asymptotic theory. The method is applied to estimate insolvent probability and to obtain higher accuracy ratio than a previous study. Supplementary materials for this article are available online.