| 摘要: |
| 本文考虑高维线性模型中的变量选择和参数估计.提出了一种广义的SELO方法求解惩罚最小二乘问题.一种坐标下降算法结合调节参数的一种连续化策略和高维BIC被用来计算相应的GSELO-PLS估计.模拟研究和实际数据分析显示了提出方法的良好表现. |
| 关键词: 连续化策略 坐标下降 高维BIC 局部线性逼近 惩罚最小二乘 |
| DOI： |
| 分类号:O212.1 |
| 基金项目:Supported by National Natural Science Foundation of China (11501578; 11501579; 11701571; 41572315); Fundamental Research Funds for the Central Universities, China University of Geosciences (Wuhan) (CUGW150809). |
|
| HIGH-DIMENSIONAL VARIABLE SELECTION WITH THE GENERALIZED SELO PENALTY |
|
SHI Yue-yong1,2, CAO Yong-xiu3, YU Ji-chang3, JIAO Yu-ling3
|
|
1.School of Economics and Management, China University of Geosciences, Wuhan 430074, China;2.Center for Resources and Environmental Economic Research, China University of Geosciences, Wuhan 430074, China;3.School of Statistics and Mathematics, Zhongnan University of Economics and Law, Wuhan 430073, China
|
| Abstract: |
| In this paper, we consider the variable selection and parameter estimation in high-dimensional linear models. We propose a generalized SELO (GSELO) method for solving the penalized least-squares (PLS) problem. A coordinate descent algorithm coupled with a continuation strategy and high-dimensional BIC on the tuning parameter are used to compute corresponding GSELO-PLS estimators. Simulation studies and a real data analysis show the good performance of the proposed method. |
| Key words: continuation strategy coordinate descent high-dimensional BIC local linear approximation penalized least squares |
