Impossibility of Distribution-Free Predictive Inference for Individual Treatment Effects
Abstract
Uncertainty quantification for individual treatment effects (ITEs) is a daunting challenge in causal inference. Motivated by recent advances in conformal prediction, several works aim to construct distribution-free prediction sets for ITEs with desired coverage under standard assumptions such as strong ignorability and overlap. In this paper, we show that such goals are fundamentally unattainable. Specifically, we establish finite-sample and asymptotic impossibility results demonstrating that any distribution-free prediction set achieving desired coverage for ITEs must be trivial, in the sense that it has infinite expected length. This infinite-expected-length conclusion requires no regularity assumption on the covariates and holds even in their absence. Our analysis builds on ideas from the hardness of conditional independence testing and exploits the missing-data nature of causal inference. These results provide a new perspective on existing methods, clarifying that their apparent success necessarily relies on additional structural assumptions beyond standard causal assumptions.
Related
- Conformal ordering: Distribution-free predictive inference for orders and rank
- Distribution-free predictive inference with instrumental variables
- A Shape-Adaptive Gamma-Mixture Power Shanker Distribution: Boundary-Aware Inference, Censored Regression, and Predictive Validation
- Sharp Design-based Inference for Quantiles of Individual Treatment Effects
- Epanechnikov Pearson Type V distribution for reliability and predictive maintenance modeling
BibTeX
@misc{tao2026impossibility,
title = {{Impossibility of Distribution-Free Predictive Inference for Individual Treatment Effects}},
author = {Chongguang Tao and Zheng Zhou and Yuhong Yang},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2605.05051},
eprint = {2605.05051},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.05051},
}
RIS
TY - UNPB
TI - Impossibility of Distribution-Free Predictive Inference for Individual Treatment Effects
AU - Chongguang Tao
AU - Zheng Zhou
AU - Yuhong Yang
PY - 2026
DA - 2026/11/04
PB - arXiv
DO - 10.48550/arxiv.2605.05051
UR - https://arxiv.org/abs/2605.05051
AB - Uncertainty quantification for individual treatment effects (ITEs) is a daunting challenge in causal inference. Motivated by recent advances in conformal prediction, several works aim to construct distribution-free prediction sets for ITEs with desired coverage under standard assumptions such as strong ignorability and overlap. In this paper, we show that such goals are fundamentally unattainable. Specifically, we establish finite-sample and asymptotic impossibility results demonstrating that any distribution-free prediction set achieving desired coverage for ITEs must be trivial, in the sense that it has infinite expected length. This infinite-expected-length conclusion requires no regularity assumption on the covariates and holds even in their absence. Our analysis builds on ideas from the hardness of conditional independence testing and exploits the missing-data nature of causal inference. These results provide a new perspective on existing methods, clarifying that their apparent success necessarily relies on additional structural assumptions beyond standard causal assumptions.
ER -