Tagged “confounding”

2 articles.

14 September 20263 min readCausal Inference

The Control Variable That Invents a Relationship

Two independent causes and one common effect. Adjust for the effect and the causes acquire a correlation of exactly -1: a regression of A on B recovers a coefficient of +0.0030, and adding the common effect as a control turns it into -1.0000. Selecting a sample does the same thing invisibly, which is why "control for everything you measured" is not a defensible rule.

Statistics

11 September 20268 min readCausal Inference

The Treatment That Helps Everyone and Harms the Average

A treatment that raises recovery by exactly five points in every subgroup while appearing to lower it overall, why more data makes that conclusion more confident rather than more correct, what randomisation buys that adjustment cannot, and the case where controlling for a variable manufactures an association from nothing.

StatisticsMachine Learning