Research

Publications & working papers

Work on distributional treatment effects, causal inference and machine learning. Each entry links to the working paper / preprint, the published version, and code where available.

Publications

Distributional treatment effects of content promotion: evidence from an ABEMA field experiment

with Shota Yasui, Tatsushi Oka and Yuki Oishi

The Japanese Economic Review · 2026

Working paper (arXiv)

Regression Adjustment for Estimating Distributional Treatment Effects in Randomized Controlled Trials

with Tatsushi Oka, Shota Yasui and Yuta Hayakawa

Econometric Reviews · 2026

Working paper (arXiv) Code (R)

Beyond the Average: Distributional Causal Inference under Imperfect Compliance

with Tomu Hirata, Tatsushi Oka and Shota Yasui

Neural Information Processing Systems (NeurIPS) · 2025

Working paper (arXiv) Code (Python)

On Efficient Estimation of Distributional Treatment Effects under Covariate-Adaptive Randomization

with Tomu Hirata, Tatsushi Oka and Shota Yasui

International Conference on Machine Learning (ICML) · 2025

Working paper (arXiv) Code (Python/R)

Estimating Distributional Treatment Effects in Randomized Experiments: Machine Learning for Variance Reduction

with Tatsushi Oka and Shota Yasui

International Conference on Machine Learning (ICML) · 2024

Published version Working paper (arXiv) Code (R)

Changing Preferences: An Experiment and Estimation of Market-Incentive Effects on Altruism

with Albert Ma and Daniel Wiesen

Journal of Health Economics, Vol. 92 · December 2023, 102808

Published version Working paper (SSRN)

Working papers

Efficient and Scalable Estimation of Distributional Treatment Effects with Multi-Task Neural Networks

with Tomu Hirata, Tatsushi Oka, Shota Yasui and Shingo Uto

Working paper (arXiv)

The Heterogeneous Impact of Changes in Default Gift Amounts on Fundraising

with Susan Athey, Matias Cersosimo, Kristine Koutout and Shanjukta Nath

Working paper (SSRN)

Contextual Bandits in a Survey Experiment on Charitable Giving: Within-Experiment Outcomes versus Policy Learning

with Susan Athey, Vitor Hadad, Sanath Krishnamurthy, Weiwen Leung and Joseph Jay Williams

Working paper (arXiv) Slides

Identification and Inference for Welfare Gains without Unconfoundedness

Working paper (arXiv)

Software

dte-adj — Distributional Treatment Effects and Regression Adjustment

A Python package for estimating distributional treatment effects with regression adjustment.

PyPI GitHub