Transformer-Based Spatial-Temporal Counterfactual Outcomes Estimation

TL;DR

This paper introduces a Transformer-based framework for estimating counterfactual outcomes in spatial-temporal data, improving accuracy and generalization over classical models, validated through simulations and real-world forest loss analysis.

Contribution

It presents a novel Transformer-based estimator for spatial-temporal counterfactual outcomes that is consistent, asymptotically normal, and outperforms baseline methods.

Findings

Stronger estimation capability than baseline methods

Effective in real-world forest loss analysis in Colombia

Estimator is consistent and asymptotically normal

Abstract

The real world naturally has dimensions of time and space. Therefore, estimating the counterfactual outcomes with spatial-temporal attributes is a crucial problem. However, previous methods are based on classical statistical models, which still have limitations in performance and generalization. This paper proposes a novel framework for estimating counterfactual outcomes with spatial-temporal attributes using the Transformer, exhibiting stronger estimation ability. Under mild assumptions, the proposed estimator within this framework is consistent and asymptotically normal. To validate the effectiveness of our approach, we conduct simulation experiments and real data experiments. Simulation experiments show that our estimator has a stronger estimation capability than baseline methods. Real data experiments provide a valuable conclusion to the causal effect of conflicts on forest loss in Colombia. The source code is available at https://github.com/lihe-maxsize/DeppSTCI_Release_Version-master.