Munich Personal RePEc Archive

Shapley value regression and the resolution of multicollinearity

Mishra, SK (2016): Shapley value regression and the resolution of multicollinearity.

[img]
Preview
PDF
MPRA_paper_72116.pdf

Download (198kB) | Preview

Abstract

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.

UB_LMU-Logo
MPRA is a RePEc service hosted by
the Munich University Library in Germany.