Timezone: »

 
Learning Treatment Effects in Panels with General Intervention Patterns
Vivek Farias · Andrew Li · Tianyi Peng

@
in Causal Inference Challenges in Sequential Decision Making: Bridging Theory and Practice »
The problem of causal inference with panel data is a central econometric question. The following is a fundamental version of this problem: Let $M^*$ be a low-rank matrix and $E$ be a zero-mean noise matrix. For a `treatment' matrix $Z$ with entries in $\{0,1\}$ we observe the matrix $O$ with entries $O_{ij} := M^*_{ij} + E_{ij} + \mathcal{T}_{ij} Z_{ij}$ where $\mathcal{T}_{ij} $ are unknown, heterogeneous treatment effects. The problem requires we estimate the average treatment effect $\tau^* := \sum_{ij} \mathcal{T}_{ij} Z_{ij} / \sum_{ij} Z_{ij}$. The synthetic control paradigm provides an approach to estimating $\tau^*$ when $Z$ places support on a single row. This paper extends that framework to allow rate-optimal recovery of $\tau^*$ for general $Z$, thus broadly expanding its applicability. Our guarantees are the first of their type in this general setting. Computational experiments on synthetic and real-world data show a substantial advantage over competing estimators.

Author Information

Vivek Farias (Massachusetts Institute of Technology)
Andrew Li (Carnegie Mellon University)
Tianyi Peng (Massachusetts Institute of Technology)

More from the Same Authors