| 相加模型中成分函数选择的新方法 |
| |
| 引用本文： | 王彬寰,邓金兰,陈国迪.相加模型中成分函数选择的新方法[J].四川大学学报(自然科学版),2009,46(1):31-36. |
| |
| 作者姓名： | 王彬寰 邓金兰 陈国迪 |
| |
| 作者单位： | 四川大学数学学院;四川大学数学学院;四川大学数学学院 |
| |
| 基金项目： | 国家自然科学基金（10771148） |
| |
| 摘 要： | 作者提出一种新的相加模型中成分函数选择方法，并将Tibshirani 的LASSO方法推广到非参数的相加模型.数值模拟研究显示该方法可以同时估计与选择成分函数，计算简便.
|
| 关 键 词： | 样条 LASSO Adaptive Ridge 相加模型 成分函数 变量选择 判罚 |
A new method for components selection in additive models |
| |
| WANG Bin-Huan,DENG Jin-Lan and Chen Guo-Di.A new method for components selection in additive models[J].Journal of Sichuan University (Natural Science Edition),2009,46(1):31-36. |
| |
| Authors: | WANG Bin-Huan DENG Jin-Lan and Chen Guo-Di |
| |
| Institution: | School of Mathematics, Sichuan University;School of Mathematics, Sichuan University;School of Mathematics, Sichuan University |
| |
| Abstract: | A new method for components selection, especially designed for additive models, is proposed. Additionally,the authors generalize LASSO proposed by Tibshirani to nonparametric additive models. The simulation results show that the approach could estimate and select component functions simultaneously, and the computation is convenient. |
| |
| Keywords: | splines LASSO adaptive ridge variable selection additive model penalization |
|
| 点击此处可从《四川大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《四川大学学报(自然科学版)》下载免费的PDF全文 |
