| 响应变量存在缺失时部分线性模型的经验似然推断 |
| |
| 引用本文： | 杨宜平,薛留根,程维虎.响应变量存在缺失时部分线性模型的经验似然推断[J].高校应用数学学报(A辑),2010,25(1). |
| |
| 作者姓名： | 杨宜平 薛留根 程维虎 |
| |
| 作者单位： | 北京工业大学应用数理学院,北京,100124 |
| |
| 基金项目： | 国家自然科学基金，高等学校博士学科点专项科研基金，北京市自然科学基金，北京市属高等学校人才强教计划资助项目 |
| |
| 摘 要： | 考虑响应变量带有缺失的部分线性模型,采用借补的思想,研究了参数部分和非参数部分的经验似然推断,证明了所提出的经验对数似然比统计量依分布收敛到χ2分布,由此构造参数部分和函数部分的置信域和逐点置信区间.对参数部分,模拟比较了经验似然与正态逼近方法;对函数部分,模拟了函数的逐点置信区间.
|
| 关 键 词： | 部分线性模型 借补 经验似然 缺失数据 置信区间 |
Empirical likelihood for partially linear models with missing response variables |
| |
| YANG Yi-ping,XUE Liu-gen,CHENG Wei-hu.Empirical likelihood for partially linear models with missing response variables[J].Applied Mathematics A Journal of Chinese Universities,2010,25(1). |
| |
| Authors: | YANG Yi-ping XUE Liu-gen CHENG Wei-hu |
| |
| Abstract: | A partially linear model with missing response variables is considered. A imputation approach is developed. Empirical likelihood inference for the regression coefficients and the nonpara-metric function is investigated. The proposed empirical-likelihood-based stati8tics are shown to have chi-squared limits. For the regression coefficients, a simulation study is conducted to compare the performance of the empirical likelihood method and the normal approximation-based method. For the nonparametric function, the pointwise confidence intervals are calculated. |
| |
| Keywords: | partially linear model imputation empirical likelihood missing data confidenceregion |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
