Shapley Value Regression and the Resolution of Multicollinearity
Shapley Value Regression and the Resolution of Multicollinearity
复制标题
Shapley 值回归和多重共线性的解决
DOI：
--
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
Sudhanshu K. Mishra
中科院分区：
文献类型：
--
作者：
Sudhanshu K. Mishra
经验数据中的多重共线性违反了线性回归模型中回归量之间的独立性假设，这通常导致无法拒绝错误的零假设。它也可能给系数分配错误的符号。Shapley值回归也许是解决这个问题的最好方法。本文简化了R2的Shapley值分解算法，并给出了执行该算法的计算机程序。然而，当回归变量的数量超过10个时，Shapley值回归变得越来越不可行，尽管在实践中，一个好的回归模型可能不超过10个回归变量。
Multicollinearity in empirical data violates the assumption of independence among the regressors in a linear regression model that often leads to failure in rejecting a false null hypothesis. It also may assign wrong sign to coefficients. Shapley value regression is perhaps the best methods to combat this problem. The present paper simplifies the algorithm of Shapley value decomposition of R2 and provides a computer program that executes it. However, Shapley value regression becomes increasingly impracticable as the number of regressor variables exceeds 10, although, in practice, a good regression model may not have more than ten regressors.