Rethinking Distributional IVs: KAN-Powered D-IV-LATE & Model Choice

TL;DR

This paper shows that the choice of machine learning models for nuisance functions in causal inference significantly affects results, comparing standard models with a novel KAN-based estimator in a distributional IV context.

Contribution

It introduces a new KAN-powered D-IV-LATE estimator and demonstrates its asymptotic properties and impact on substantive conclusions.

Findings

KAN-based estimator reveals more complex treatment heterogeneity

Model choice can alter causal inference outcomes

Empirical analysis highlights importance of model selection

Abstract

The double/debiased machine learning (DML) framework has become a cornerstone of modern causal inference, allowing researchers to utilise flexible machine learning models for the estimation of nuisance functions without introducing first-order bias into the final parameter estimate. However, the choice of machine learning model for the nuisance functions is often treated as a minor implementation detail. In this paper, we argue that this choice can have a profound impact on the substantive conclusions of the analysis. We demonstrate this by presenting and comparing two distinct Distributional Instrumental Variable Local Average Treatment Effect (D-IV-LATE) estimators. The first estimator leverages standard machine learning models like Random Forests for nuisance function estimation, while the second is a novel estimator employing Kolmogorov-Arnold Networks (KANs). We establish the asymptotic properties of these estimators and evaluate their performance through Monte Carlo simulations. An empirical application analysing the distributional effects of 401(k) participation on net financial assets reveals that the choice of machine learning model for nuisance functions can significantly alter substantive conclusions, with the KAN-based estimator suggesting more complex treatment effect heterogeneity. These findings underscore a critical "caveat emptor". The selection of nuisance function estimators is not a mere implementation detail. Instead, it is a pivotal choice that can profoundly impact research outcomes in causal inference.