Identification and Estimation of Long-Term Treatment Effects with Monotone Missing
This work addresses a gap in causal inference for domains like healthcare or social sciences where long-term outcomes are hard to collect due to monotone missingness, though it is incremental in extending existing methods to this specific data issue.
The paper tackles the problem of estimating long-term treatment effects under monotone missing data, where outcomes are missing in a sequential pattern, by introducing a sequential missingness assumption and proposing novel estimation methods including BalanceNet to improve stability. Experiments on benchmark datasets show the methods are effective, with concrete performance gains reported.
Estimating long-term treatment effects has a wide range of applications in various domains. A key feature in this context is that collecting long-term outcomes typically involves a multi-stage process and is subject to monotone missing, where individuals missing at an earlier stage remain missing at subsequent stages. Despite its prevalence, monotone missing has been rarely explored in previous studies on estimating long-term treatment effects. In this paper, we address this gap by introducing the sequential missingness assumption for identification. We propose three novel estimation methods, including inverse probability weighting, sequential regression imputation, and sequential marginal structural model (SeqMSM). Considering that the SeqMSM method may suffer from high variance due to severe data sparsity caused by monotone missing, we further propose a novel balancing-enhanced approach, BalanceNet, to improve the stability and accuracy of the estimation methods. Extensive experiments on two widely used benchmark datasets demonstrate the effectiveness of our proposed methods.