×

Robust neural network with applications to credit portfolio data analysis. (English) Zbl 1245.91099

Summary: We study nonparametric conditional quantile estimation via neural network structure. We propose an estimation method that combines quantile regression and neural network (robust neural network, RNN). It provides good smoothing performance in the presence of outliers and can be used to construct prediction bands. A Majorization-Minimization (MM) algorithm was developed for optimization. Monte Carlo simulation study is conducted to assess the performance of RNN. Comparison with other nonparametric regression methods (e.g., local linear regression and regression splines) in real data application demonstrate the advantage of the newly proposed procedure.

MSC:

91G60 Numerical methods (including Monte Carlo methods)
62G35 Nonparametric robustness
62P05 Applications of statistics to actuarial sciences and financial mathematics
65C05 Monte Carlo methods
91G40 Credit risk
91G10 Portfolio theory
PDFBibTeX XMLCite
Full Text: DOI