| 基于最小一乘准则的最优回归模型研究 |
| |
| 引用本文： | 肖健华,林健,孙德山.基于最小一乘准则的最优回归模型研究[J].辽宁师范大学学报(自然科学版),2005,28(2):129-131. |
| |
| 作者姓名： | 肖健华 林健 孙德山 |
| |
| 作者单位： | 五邑大学,管理学院,广东,江门,529020;辽宁师范大学,数学学院,辽宁,大连,116029 |
| |
| 基金项目： | 国家自然科学基金资助项目(70471074)，广东省自然科学基金资助项目(032353) |
| |
| 摘 要： | 针对目前回归函数中普遍存在的泛化能力得不到保证的缺点，结合统计学习理论的研究成果，提出了基于最小一乘准则的最优回归模型(简称为LaOR模型)．与以往模型相比较，LaOR模型综合考虑了回归误差和置信范围，可望有效地降低回归函数的期望风险．最后，以广东省江门市为例，将LaOR应用到区域经济发展的短期预测中，取得了可以接受的预测效果．
|
| 关 键 词： | 统计学习理论 最小一乘准则 多元回归 经济预测 |
| 文章编号： | 1000-1735(2005)02-0129-03 |
| 修稿时间： | 2004年11月15 |
Research of Optimal Regress Model Based on the Least-absolute Criteria |
| |
| XIAO Jian-hua,LIN Jian,SUN De-shan.Research of Optimal Regress Model Based on the Least-absolute Criteria[J].Journal of Liaoning Normal University(Natural Science Edition),2005,28(2):129-131. |
| |
| Authors: | XIAO Jian-hua LIN Jian SUN De-shan |
| |
| Institution: | XIAO Jian-hua~1,LIN Jian~1,SUN De-shan~2 |
| |
| Abstract: | Aimming at the disadvantage of weak generalization ability that exists in most of the current regression functions,combining with the research achievement of statistic learning theory,we proposed the optimal regress model based on least-absolute criteria, or LaOR model. Compared with other regress models,LaOR model has taken regress error and confidence interval into account synthetically.In the end, LaOR model can reduce the expected risk of regress model effectively.Finally, we take the economic forecasting of Jiangmen Guangdong as an example to examine the validity of the LaOR model. |
| |
| Keywords: | statistic learning theory least-absolute criteria multiple regression economic forecasting |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
