我国城镇居民消费结构沿收入梯度的响应——基于对应分析  

Response of the Consumption Structure of Chinese Urban Residents along Income Gradient Based on Correspondence Analysis

作　　者：徐海云[1] 涂雄苓[2] 罗付岩[2] 

机构地区：[1]中南财经政法大学信息学院,湖北武汉430060 [2]桂林工学院数理系,广西桂林541004

出　　处：《系统工程》2007年第12期42-46,共5页Systems Engineering

基　　金：国家自然科学基金资助项目(10661006)

摘　　要：应用多元统计方法之一的对应分析探索了随着收入变化,城市居民消费结构的变化规律,结果显示:(a)处于不同收入梯度的人群在二维消费支出坐标系下分化严重;(b)二维空间也足以表达不同组消费结构差异的绝大部分信息;(c)各收入人群与相应的消费支出有显著而稳定的"亲缘"关系;(d)对应分析在探索多维数据内在联系具有高效性。最后,对产生这种现象的成因进行详细的分析、讨论。This paper applies the correspondence analysis （CA）, one of the multivariate statistical methods, to study the change rule of urban residents＇ consumption structure along with the variation of their income. The results show that （a） the crowds in different age gradients vary seriously in the 2D coordinate axis of consumption expense; （b） it is enough for the 2D space to explain the most of the consumption structure difference among different consumers; （c） there exists significant and stable neighborhood between various consumption expenditures and the related income gradients; （d） correspondence analysis can analyze efficiently the inherent relation of multi-groups data in multi-dimenslons. Finally, the causes to these phenomena have been analyzed and discussed concretely.

关 键 词：多元统计 降维 R-因子分析 Q-因子分析 

分 类 号：F124.7[经济管理—世界经济] TU984[建筑科学—城市规划与设计]