发帖
雷达卡
A Proxy Variable View of Shared Confounding
Yixin Wang 1 David M. Blei 2
Abstract confounder) are difficult to obtain, e.g., requiring expensive
Causal inference from observational data real-time tracking of activities. Lifestyle is necessarily an
can be biased by unobserved confounders. unobserved confounder.
Confounders—the variables that affect both the Here we focus on the setting where multiple t ...
京公网安备 11010802022788号
