A0261
Title: Continuous treatment effects with surrogate outcomes
Authors: Zhenghao Zeng - Carnegie Mellon University (United States) [presenting]
David Arbour - Adobe Research (United States)
Avi Feller - University of California at Berkeley (United States)
Raghavendra Addanki - Adobe Research (United States)
Ryan Rossi - Adobe Research (United States)
Ritwik Sinha - Adobe Research (United States)
Edward Kennedy - Carnegie Mellon University (United States)
Abstract: In many real-world causal inference applications, the primary outcomes (labels) are often partially missing, especially if they are expensive or difficult to collect. If the missingness depends on covariates (i.e., missingness is not completely at random), analyses based on fully observed samples alone may be biased. Incorporating surrogates, which are fully observed post-treatment variables related to the primary outcome, can improve estimation in this case. The role of surrogates is studied in estimating continuous treatment effects and propose a doubly robust method to efficiently incorporate surrogates in the analysis, which uses both labeled and unlabeled data and does not suffer from the above selection bias problem. Importantly, the asymptotic normality of the proposed estimator is established, and possible improvements in the variance are shown compared with methods that solely use labeled data. Extensive simulations show the methods enjoy appealing empirical performance.
