首页 | 本学科首页   官方微博 | 高级检索  
检索        

多母体稳健Fisher判别方法及其Monte Carlo模拟
引用本文:颜光宇,夏结来.多母体稳健Fisher判别方法及其Monte Carlo模拟[J].数理医药学杂志,1993,6(3):21-26.
作者姓名:颜光宇  夏结来
作者单位:第一军医大学军队卫生统计教研室,第四军医大学卫生统计教研室 广州 510515,西安 710032
基金项目:国家自然科学基金(编号 39070765)
摘    要:传统Fisher判别方法不具稳健性,其分析结果易受异常值的干扰。本文采用近年新兴的投影追踪(Projection-Pursuit)的思想,将高维数据降维而估计出稳健的协差阵,进而建立一种新的能抗异常值干扰的稳健的Fisher判别方法。对该方法进行了计算机Monte Carlo模拟比较,结果显示当异常值存在时,该法优于传统的Fisher判别方法。

关 键 词:判别分析  投影追踪  稳健统计

A New Robust Fisher Discrimination Procedure and Its Monte Carlo simulation
Yan Guangyu,et al..A New Robust Fisher Discrimination Procedure and Its Monte Carlo simulation[J].Journal of Mathematical Medicine,1993,6(3):21-26.
Authors:Yan Guangyu  
Institution:Yan Guangyu,et al . Department of Medical Statistics First Military Medical University Guangzhou 510515
Abstract:Fisher Discrimination is a important method in the aera of multivariate analysis, and it has being widely used in biomedicine, geology, economics and other scientific aeras. But. it is nonrobust and can easily be influenced by outliers. In this paper, we developed a new robust Fisher discrimination procedure via the technique of Projection-Pursuit. Results of Monte Carlo simulation and practical examples showed that the new procedure wins advantages over the classical Fisher discrimination when some outliers exist in data set.
Keywords:Discriminant analysis  Projection-Pursuit  Robust statistics
本文献已被 CNKI 维普 等数据库收录!
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