首页 | 本学科首页   官方微博 | 高级检索  
     检索      

生长曲线模型的变量选择
引用本文:高采文,朱晓琳,曾林蕊.生长曲线模型的变量选择[J].应用概率统计,2014,30(2):213-222.
作者姓名:高采文  朱晓琳  曾林蕊
作者单位:山西大同大学数学与计算机科学学院, 华东师范大学金融与统计学院
摘    要:生长曲线模型是一个典型的多元线性模型, 在现代统计学上占有重要地位. 文章首先基于Potthoff-Roy变换后的生长曲线模型, 采用自适应LASSO为惩罚函数给出了参数矩阵的惩罚最小二乘估计, 实现了变量的选择. 其次, 基于局部渐近二次估计, 对生长曲线模型的惩罚最小二乘估计给出了统一的近似估计表达式. 接着, 讨论了经过Potthoff-Roy变换后模型的惩罚最小二乘估计, 证明了自适应LASSO具有Oracle性质. 最后对几种变量选择方法进行了数据模拟. 结果表明自适应LASSO效果比较好. 另外, 综合考虑, Potthoff-Roy变换优于拉直变换.

关 键 词:变量选择  Potthoff-Roy变换  Oracle性质  自适应LASSO.

The Variable Selection of the Growth Curve Model
Gao Caiwen,Zhu Xiaolin,Zeng Linrui.The Variable Selection of the Growth Curve Model[J].Chinese Journal of Applied Probability and Statisties,2014,30(2):213-222.
Authors:Gao Caiwen  Zhu Xiaolin  Zeng Linrui
Institution:School of Mathematics and Computre Science, Shanxi Datong University; School of Finance and Statistics, East China Normal University
Abstract:Growth curve model is a general multivariable linear model. It plays an important role in modern statistics. In this paper, firstly, we define the penalized least squares for growth curve model, after transforming it by the Potthoff-Roy transformation. By using adaptive LASSO we can get corresponding estimation, as well as achieve the variable selection. Then, the penalized least squares estimation of the growth curve model is presented with a unified expression of approximate estimation. In addition, we discuss the properties of the penalized least squares estimations of the growth curve model, which is transformed by Potthoff-Roy transformation, and the properties, which are Oracle properties, are proved in this paper. By using the criteria to measure estimation and variable selection, we compare several penalized least squares estimations and the effect of variable selection of different penalty functions. The result shows that the adaptive LASSO performs better in parameter estimation and variable selection. Besides, we compare different transformations. Results indicate that Potthoff-Roy transformation performs better than matrix stacking transformation when considering variable selection and parameter estimation comprehensively.
Keywords:
点击此处可从《应用概率统计》浏览原始摘要信息
点击此处可从《应用概率统计》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号