Robust neural network with applications to credit portfolio data analysis.
Robust neural network with applications to credit portfolio data analysis.
复制标题
DOI：
10.4310/sii.2010.v3.n4.a2
复制
发表时间：
2010
影响因子：
0.8
通讯作者：
Zhang Y
中科院分区：
文献类型：
--
作者：
Feng Y;Li R;Sudjianto A;Zhang Y
本文利用神经网络结构研究了非参数条件分位数估计。提出了一种分位数回归与神经网络（Robust Neural Network，RNN）相结合的估计方法。它在存在离群值的情况下提供了良好的平滑性能，并可用于构建预测带。提出了一种优化的多变量最小化（MM）算法。蒙特卡洛模拟研究进行评估的RNN的性能。与其他非参数回归方法的比较（例如，局部线性回归和回归样条）在真实的数据应用中的应用证明了新提出的过程的优点。
In this article, we study nonparametric conditional quantile estimation via neural network structure. We proposed 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.