Fast and Efficient Estimation of Generalized Additive Partially Linear Model

• Activity Number: 477
• Type: Topic Contributed
• Date/Time: Wednesday, August 7, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Rong Liu*+
• Companies: University of Toledo
• Keywords: Bandwidths ; B spline ; knots ; mixing ; credit ; link function

Abstract

One central field of modern financial risk management is corporate credit rating in which default prediction plays a vital role. The hybrid spline-backfitted kernel estimation method has been used for generalized additive model (GAM) in Liu et al. (2013) to make default prediction. This method combines the best features of both spline and kernel methods for making fast, efficient and reliable inference on component functions. We extend this method to the more flexible generalized additive partially linear model (GAPLM), which contains not only additive nonparametric component functions but also parametric parts. The asymptotic normal distribution and uniform convergence rate are obtained and simulation results support the theoretical properties. The method is applied to estimate insolvent probability and obtain higher accuracy ratio than previous study.