TabCF: Distributional Control Function Estimation with Tabular Foundation Models

TL;DR

TabCF is a new method leveraging tabular foundation models for distributional causal inference, offering accurate, fast, and easy-to-tune estimation of effects like means and quantiles in various data scenarios.

Contribution

It introduces TabCF, a simple control function regression approach using foundation models for distributional causal effect estimation, with a copula-based extension for multivariate outcomes.

Findings

TabCF outperforms existing methods on synthetic and real datasets.

It enables fast and transparent estimation of distributional effects.

The approach is effective for small to medium-sized data scenarios.

Abstract

Instrumental variable (IV) and control function (CF) methods are powerful tools for causal effect estimation in the presence of unmeasured confounding, yet most existing approaches target only mean effects and/or demand substantial fitting and tuning effort. In this paper, we introduce a simple method, TabCF, for control function regression using tabular foundation models, which enables accurate, fast, identification-transparent, and tuning-light causal estimation of distributional quantities, such as interventional means and quantiles; we also propose a copula-based approximation for multivariate outcomes. TabCF performs favorably against representative methods across a broad range of small- to medium-sized synthetic and real data scenarios. The central message is two-fold: for practitioners, it highlights that TabCF is an effective tool for distributional causal inference; for researchers, it suggests that the proposed approach could be considered a strong baseline for future method development. Code is available at https://github.com/GepingChen/TabCF.