|国家预印本平台
首页|多维系统优化中马氏田口强相关问题研究

多维系统优化中马氏田口强相关问题研究

Multicollinearity Analysis when using Mahalanobis-Taghchi System for Multidimensional System Optimization

中文摘要英文摘要

马氏田口是研究多维系统优化与诊断的一种新方法。然而,强相关问题的存在使得马氏距离难以计算或很不准确,给马氏田口的应用带来了问题。对传统马氏田口方法进行了介绍,分析了强相关问题对多维系统马氏田口优化的影响,在此基础上提出了一种解决强相关问题的新方法--马氏田口M-P广义逆矩阵法。相对于传统的马氏田口逆矩阵法,马氏田口M-P广义逆矩阵法具有存在性、唯一性和通用性的优点,能更有效地应用于多维系统的优化实践。通过对某医院血粘度诊断系统的优化与分析,进一步证实了马氏田口M-P广义逆矩阵的有效性。

Mahalanobis-Taguchi System (MTS) is a new method for multidimensional system analysis. However, the calculation of Mahalanobis distance becomes difficult and inaccurate because of multicollinearity, which limits the application of MTS. Traditional MTS method is introduced and the effect on MTS by multicollinearity is analyzed; a new method, M-P generalized inverse matrix method of MTS, is put forward. Compared with traditional MTS method, the new method has three advantages: existence, uniqueness and commonality, which can be more effectively used for multidimensional system optimization. And a blood-viscosity-diagnose system is optimized and analyzed to illustrate the effectiveness of the proposed method.

何桢、韩亚娟、宋国防

医学研究方法数学

多维系统优化马氏田口强相关问题M-P广义逆矩阵法

multidimensional system optimizationMahalanobis-Taguchi SystemmulticollinearityM-P generalized inverse matrix

何桢,韩亚娟,宋国防.多维系统优化中马氏田口强相关问题研究[EB/OL].(2011-10-26)[2025-07-21].http://www.paper.edu.cn/releasepaper/content/201110-290.点此复制

评论