基于多元统计分析方法对我国不同省市自治区的发展水平初步分析  

A Preliminary Analysis on the Development Level of Different Regions in China Based on Multivariate Statistical Analysis

作　　者：李华[1] 石荣平 

机构地区：[1]长春大学理学院,长春130022

出　　处：《长春大学学报》2015年第10期55-59,共5页Journal of Changchun University

摘　　要：将多元统计分析方法与实际应用结合对我国不同省市自治区的发展水平进行初步分析。多元统计方法涉及较为复杂的数学理论,内容众多,本文选取其中三方面内容作为分析工具:通过典型相关分析找出几对主要综合变量,研究不同因子的相关关系;通过聚类分析对我国不同省市自治区进行分类并探究其原因;通过主成分分析以及结合聚类分析对我国不同省市自治区作综合评价。With the combination of multivariate statistical analysis and practical application, this paper makes a preliminary analysis on the development level of different regions in China. Multivariate statistical analysis with rich contents concerns complicated mathematical theories, three of which are selected as analysis tools, including that canonical correlation analysis is used to find out a few main comprehensive variables and study the relationship among different factors; clustering analysis is used to classify the different regions in China and explore the corresponding reasons; principal component analysis is used to provide the comprehensive evaluation.

关 键 词：指标 典型相关分析 聚类分析 主成分分析 

分 类 号：O212.4[理学—概率论与数理统计]