摘要
研究半参数部分线性变系数模型的有偏估计,当回归模型参数部分自变量存在多重共线性时,在随机线性约束条件下,融合Profile最小二乘估计、加权混合估计和Liu估计构造回归模型参数分量改进的加权混合Profile-Liu估计,并在一定正则条件下证明估计量的渐近性质,最后利用蒙特卡洛数值模拟验证所提出估计量的有限样本表现性.
In this paper,we investigate the biased estimation of semiparametric partially linear varying-coefficient model under stochastic linear constraints,when the parametric part of the regression model in the presence of multicollinearity.An Improved weighted mixed Profile-Liu estimation for the parametric part is constructed based on profile least squares estimation,weighted mixed estimation and Liu estimation,and the asymptotic properties of the estimators are proved under certain regular conditions.Finally,a Monte Carlo simulation study is conducted to verify the finite sample performance of the proposed estimators.
作者
张巍巍
ZHANG Wei-wei(College of Science,Inner Mongolia Agricultural University,Hohhot 010018,China)
出处
《数学的实践与认识》
2021年第3期128-135,共8页
Mathematics in Practice and Theory
基金
内蒙古农业大学基础学科科研启动基金项目(JC2017002)
内蒙古自治区高等学校科学研究项目(NJZY19045)。
关键词
半参数部分线性变系数模型
随机线性约束
Profile最小二乘估计
Liu估计
加权混合估计
渐近性质
semiparametric partially linear varying-coefficient model
stochastic linear constraints
profile least squares estimation
Liu estimation
weighted mixed estimation
asymptotic properties