成分数据相关系数的计算方法

Computing Correlation Coefficient of Compositional Data

成分数据具有非常复杂的数学性质,很多传统的统计分析方法对其是失效的,因此,在研究中必须采用特殊处理和专门技术.着重讨论了成分数据相关系数的计算方法,由于普通数据的相关系数计算方法只适用于两组单变量数据,而传统的典型相关分析又鉴于成分数据的特殊性质而不能直接使用,故结合logratio变换和典型相关分析技术,提出了一种针对成分数据的相关系数计算方法,成功地解决了这一问题.

Many traditional statistical methods are not available to compositional data for their complex mathematical characteristics. Therefore special tools should be used in the analysis of related problems. This paper focuses on the correlation coefficient of compositional data. Since the algorithm of correlation coefficient of common data is not adaptive to multi-variables data, and traditional canonical correlation analysis cannot be directly applied to compositional data, a new approach is presented which combines logratio transformation and canonical correlation analysis, and succeeds in computing correlation coefficient of compositional data.