HIERARCHICAL PARTITIONING

被引:1022
作者
CHEVAN, A [1 ]
SUTHERLAND, M [1 ]
机构
[1] UNIV MASSACHUSETTS,CTR STAT CONSULTING,AMHERST,MA 01003
关键词
DECOMPOSITION; REGRESSION;
D O I
10.2307/2684366
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Many users of regression methods are attracted to the notion that it would be valuable to determine the relative importance of independent variables. This article demonstrates a method based on hierarchies that builds on previous efforts to decompose R2 through incremental partitioning. The standard method of incremental partitioning has been to follow one order among the many possible orders available. By taking a hierarchical approach in which all orders of variables are used, the average independent contribution of a variable is obtained and an exact partitioning results. Much the same logic is used to divide the joint effect of a variable. The method is general and applicable to all regression methods, including ordinary least squares, logistic, probit, and loglinear regression. A validation test demonstrates that the algorithm is sensitive to the relationships in the data rather than the proportion of variability accounted for by the statistical model used.
引用
收藏
页码:90 / 96
页数:7
相关论文
共 19 条