Detecting confounding in multivariate linear models via spectral analysis
Abstract
We study a model where one target variable Y is correlated with a vector X:=(X_1,...,X_d) of predictor variables being potential causes of Y. We describe a method that infers to what extent the statistical dependences between X and Y are due to the influence of X on Y and to what extent due to a hidden common cause (confounder) of X and Y. The method relies on concentration of measure results for large dimensions d and an independence assumption stating that, in the absence of confounding, the vector of regression coefficients describing the influence of each X on Y typically has `generic orientation' relative to the eigenspaces of the covariance matrix of X. For the special case of a scalar confounder we show that confounding typically spoils this generic orientation in a characteristic way that can be used to quantitatively estimate the amount of confounding.
Keywords
Cite
@article{arxiv.1704.01430,
title = {Detecting confounding in multivariate linear models via spectral analysis},
author = {Dominik Janzing and Bernhard Schoelkopf},
journal= {arXiv preprint arXiv:1704.01430},
year = {2017}
}
Comments
27 pages, 16 figures