摘要
针对“因变量和自变量都是成分数据的前提下,如何建立它们之间的线性回归”的基本问题,以经典线性回归分析法为基础,结合对称Logratio变换,建立了一种基于偏最小二乘回归分析的成分数据预测模型,并对该模型进行了理论实证分析,论证了该模型的可行性与优良性,从而为解决具有成分数据信息的多重相关变量回归问题提供新的途径.
This paper discusses the problem about how to establish the linear regression between dependent variables and indepent variables with the compositional data. Based on the traditional linear regression method and symmetrical Logratio transformation, a compositional data forecast model Based on partial least-squares regression analysis is established, then the theoretical analysis is given to show the feasibility and superiority of this model. Therefore, it gives a new way to solve the problem of mulriple correlative variables regression with the compositional data.
出处
《华中师范大学学报（自然科学版）》
CAS
CSCD
2006年第2期161-163,共3页
Journal of Central China Normal University：Natural Sciences
基金
湖北省教育厅重点项目(2004D001)
关键词
成分数据
预测模型
偏最小二乘回归分析
对称Logratio变换
compositional data
forecast model
partial least-squares regression analysis
symmetrical Logratio transformation
