| 基于典型相关的线性回归模型 |
| |
| 引用本文： | 赖艺芬,梁飞豹.基于典型相关的线性回归模型[J].福州大学学报(自然科学版),2004,32(4):438-441. |
| |
| 作者姓名： | 赖艺芬 梁飞豹 |
| |
| 作者单位： | 福州大学数学与计算机科学学院,福建,福州,350002 |
| |
| 摘 要： | 通过实验、分析、并理论证明,在满秩情况下,由典型相关求解线性回归模型与直接求解线性回归模型的结果是一致的.将该方法推广到不满秩的情况,提出了一种新的求解线性回归模型的方法,虽求解结果牺牲部分的拟合度,但在一定程度上增强了预测效果.
|
| 关 键 词： | 线性回归 典型相关 预测 理想状态 变量 |
| 文章编号： | 1000-2243(2004)04-0438-04 |
| 修稿时间： | 2004年3月31日 |
Linear regression model based on canonical correlation |
| |
| LAI Yi-fen,LIANG Fei-bao.Linear regression model based on canonical correlation[J].Journal of Fuzhou University(Natural Science Edition),2004,32(4):438-441. |
| |
| Authors: | LAI Yi-fen LIANG Fei-bao |
| |
| Institution: | (College of Mathematics and Computer Science, Fuzhou University, Fuzhou, Fujian 350002, China) |
| |
| Abstract: | In the case of full rank, we prove the consistent result between the method that solves the linear regression model via canonical correlation and the way which directly solves it. Furthermore, the paper popularizes the method for the situation of non-full rank and proposes a kind of new method to solve linear regression model. The method sacrifices partial fitness to strengthen the efficiency of forecast. |
| |
| Keywords: | linear regression canonical correlation prediction perfect state variable |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《福州大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《福州大学学报(自然科学版)》下载免费的PDF全文 |
