| 函数系数部分线性回归模型的变量选择方法 |
| |
| 引用本文： | 冯井艳,鞠思秋.函数系数部分线性回归模型的变量选择方法[J].太原师范学院学报(自然科学版),2012(4):75-77. |
| |
| 作者姓名： | 冯井艳 鞠思秋 |
| |
| 作者单位： | 山西大同大学数学系;华东师范大学金融与统计学院 |
| |
| 基金项目： | 山西大同大学科研项目(2010K4) |
| |
| 摘 要： | 函数系数部分线性回归模型是变系数模型中的一种特殊情形，文章对这种新的变系数模型的变量选择问题进行了主要研究．首先，运用局部多项式方法得到非参数项的估计，并且使用B样条逼近函数系数，选取SCAD惩罚作为变量选择方法．其次，得到了估计的渐近性质．最后，模拟说明了该估计方法较好地达到了变量选择的目的．
|
| 关 键 词： | 变量选择 B样条 SCAD惩罚函数 局部多项式 |
Variable Selection for Function-Coefficient Partially Linear Model |
| |
| Feng Jingyan,Ju Siqiu.Variable Selection for Function-Coefficient Partially Linear Model[J].Journal of Taiyuan Normal University:Natural Science Edition,2012(4):75-77. |
| |
| Authors: | Feng Jingyan Ju Siqiu |
| |
| Institution: | 1.Department of Mathematics,Shanxi Datong University,Datong 037009; 2.Department of Statistics,East China Normal University,Shanghai 200241,China) |
| |
| Abstract: | The function-coefficient partially linear model is an extension of the varying coefficient model..We mainly focus on the variable selection for the function-coefficient components. Firstly, use local polynomial to get the estimation of nonparametric, meanwhile use B- spline to fit the function coefficient and utilize SCAD as a penal factor to select variable. Second- ly,consider the properties of the estimation. Simulation results show that the proposed variable selection procedure performs well. |
| |
| Keywords: | variable selection B-spline SCAD penalty local polynomial |
| 本文献已被 CNKI 维普 等数据库收录！ |
