测量误差数据下单指标模型的非参数模拟外推估计
Non-parametric Simulation Extrapolation Estimation of Single Index Model for Measurement Error Data
王浩 1赵培信2
作者信息
- 1. 重庆工商大学数学与统计学院,重庆 400067
- 2. 重庆工商大学数学与统计学院,重庆 400067;统计智能计算与监测重庆市重点实验室,重庆 400067
- 折叠
摘要
研究了测量误差数据下一类单指标模型的估计问题.结合局部线性平滑法和核密度估计法,提出一种基于非参数模拟外推的模型估计方法.所提出的非参数模拟外推估计,不需要假定观测变量的具体分布并且不需要假定测量误差的方差已知,具有较广的适应性.在一些正则条件下,证明了非参数模拟外推方法给出的参数估计量的渐近正态性.
Abstract
The estimation problem of a single-index model under the measurement error data is studied.By combining local linear smoothing method and kernel density estimation method,a model estimation method based on non-parametric simulation extrapolation is proposed.The proposed non-parametric simulation extrapolation esti-mate does not need to assume the specific distribution of the observed variables and the variance of the measure-ment error is known,so it has a wide adaptability.Under some regular conditions,the asymptotic normality of the parameter estimators given by the non-parametric simulation extrapolation method is proven.
关键词
单指标模型/测量误差数据/核密度函数/非参数模拟外推Key words
single index model/measurement error data/kernel density function/non-parametric simulation extrapolation引用本文复制引用
出版年
2025
NETL
NSTL
万方数据