| 应用RJMCMC方法识别线性模型异常点 |
| |
| 引用本文： | 石永亮,汪四水.应用RJMCMC方法识别线性模型异常点[J].数学的实践与认识,2011,41(17). |
| |
| 作者姓名： | 石永亮 汪四水 |
| |
| 作者单位： | 苏州大学数学科学学院,江苏苏州,215006 |
| |
| 摘 要： | 传统线性模型异常点识别方法容易发生误判:正常点被归为异常点或者异常点被归为正常点.为解决此类问题,提出了应用逆跳马尔科夫蒙特卡洛方法识别异常点的思想,同时将其应用于实际数据加以检验,识别效果明显好于传统方法.
|
| 关 键 词： | 逆跳马尔科夫链蒙特卡洛方法 异常点 识别 |
Application of RJMCMC Method in the Research of Outliers in the Linear Regression Model |
| |
| SHI Yong-liang,WANG Si-shui.Application of RJMCMC Method in the Research of Outliers in the Linear Regression Model[J].Mathematics in Practice and Theory,2011,41(17). |
| |
| Authors: | SHI Yong-liang WANG Si-shui |
| |
| Institution: | SHI Yong-liang,WANG Si-shui (Mathematical Sciences,University of Soochow,Suzhou 215006,China) |
| |
| Abstract: | Traditional methods of outlier detection tend to be failure of justice:outliers are classified to be non-outliers and vice versa.Therefore a new idea about outlier detection is proposed,which is reversible jump Markov chain Monte Carlo method.With this method,outliers in a real dataset are detected and without injustice whice is just the problem traditional methods have to face with. |
| |
| Keywords: | RJMCMC outliers detection |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
