Bias Reduction for Nonparametric and Semiparametric Regression Models

Bias Reduction for Nonparametric and Semiparametric Regression Models
复制标题

非参数和半参数回归模型的偏差减少

DOI:
10.5705/ss.202017.0058
复制
发表时间:
2018-10
期刊:
影响因子:
1.4
通讯作者:
H. Peng
H. Peng
中科院分区:
数学3区
文献类型:
--
作者:
M. Cheng;T. Huang;P. Liu;H. Peng

文献摘要

参考文献

相似文献

非参数和半参数回归模型是发现响应变量和预测变量之间的非线性关系的有用的统计回归模型。然而,模型中的非参数成分的最优有效估计是有偏的,这阻碍了进一步统计推断方法的发展。本文基于局部线性拟合,提出了一种简单的非参数回归模型估计的偏差减小方法。新方法不需要使用高阶局部多项式回归来估计偏差,从而避免了高阶局部多项式拟合所带来的双带宽选择和设计稀疏性问题。它也不会夸大差异。因此,它可以很容易地应用于复杂的统计推断问题。我们扩展了我们的新方法Statistica Sinica:Newly Accepted Paper(接受版本以英文编辑为准)
Nonparametric and semiparametric regression models are useful statistical regression models to discover nonlinear relationships between the response variable and predictor variables. However, optimal efficient estimates for the nonparametric components in the models are biased which hinders the development of methods for further statistical inference. In this paper, based on the local linear fitting, we propose a simple bias reduction approach for the estimation of the nonparametric regression model. The new approach does not need to use higher-order local polynomial regression to estimate the bias, and hence avoids the double bandwidth selection and the design sparsity problems suffered by higher-order local polynomial fitting. It also does not inflate the variance. Hence it can be easily applied to complex statistical inference problems. We extend our new approach Statistica Sinica: Newly accepted Paper (accepted version subject to English editing)
DOI: 10.1214/aos/1176325632
发表时间: 1994-09-01
影响因子: 4.5
作者:
RUPPERT, D;WAND, MP
通讯作者: WAND, MP
DOI: 10.1093/biomet/85.2.333
发表时间: 1998-06
期刊: Biometrika
影响因子: 2.7
作者:
E. Choi;P. Hall
通讯作者: E. Choi;P. Hall
DOI: 10.1080/01621459.1993.10476410
发表时间: 1993-12
影响因子: 3.7
作者:
R. Eubank;P. Speckman
通讯作者: R. Eubank;P. Speckman
DOI: 10.1016/j.jmva.2010.03.003
发表时间: 2010-08-01
影响因子: 1.6
作者:
Zhang, Wenyang;Peng, Heng
通讯作者: Peng, Heng
DOI: 10.5705/ss.2012.127
发表时间: 2014
期刊: Statistica Sinica
影响因子: 1.4
作者:
Li, Gaorong;Peng, Heng;Dong, Kai;Tong, Tiejun
通讯作者: Tong, Tiejun