Beyond Imputation: Mask-Adaptive Conformal Prediction via Tree Embeddings on General Missing Data Mechanisms
Abstract
Conformal prediction (CP) provides a distribution-free framework for uncertainty quantification with finite-sample marginal coverage guarantees. Yet, missing values introduce distributional heterogeneity, where marginal coverage can hide systematic undercoverage for certain missingness patterns. At the same time, existing methods based on distances or smoothness require a continuous representation of partially observed covariates. Although imputation restores continuity, it obscures the missingness patterns and introduces task-dependent overhead. We develop a new CP framework that is \emph{adaptive to missingness patterns}, without imputing covariates. We rely on data augmentation by feature embedding that accounts for missingness. Our framework represents each input with missing values by a tree embedding together with its raw missingness mask, and then learns a regularized quantile score function. Varying the choices for the penalty function and the augmentation enables us to characterize three practically relevant validity levels: exact coordinate-wise validity, exact validity on observed masks, and approximate missingness-conditional guarantees for general mechanisms (MCAR, MAR, MNAR). Experiments on synthetic and real datasets support these guarantees and show that the proposed method reduces miscoverage risk while maintaining informative intervals, especially in non-MCAR settings. These results highlight the practical value of uncertainty quantification that adapts to missingness.
Related
- Adaptive conformal prediction on the Grassmannian
- CASCADE Conformal Prediction: Uncertainty-Adaptive Prediction Intervals for Two-Stage Clinical Decision Support
- Skew-adaptive conformal prediction
- Adaptive Conformal Prediction for Image Regression Models with Application to an Inertial Confinement Fusion Emulator
- Conformal Prediction with Stable Measurement Mechanisms
BibTeX
@misc{fan2026beyond,
title = {{Beyond Imputation: Mask-Adaptive Conformal Prediction via Tree Embeddings on General Missing Data Mechanisms}},
author = {Jiarong Fan and Juhyun Park and Thi Phuong Thuy Vo and Nicolas J-B. Brunel},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.01724},
eprint = {2611.01724},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.01724},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Beyond Imputation: Mask-Adaptive Conformal Prediction via Tree Embeddings on General Missing Data Mechanisms
AU - Jiarong Fan
AU - Juhyun Park
AU - Thi Phuong Thuy Vo
AU - Nicolas J-B. Brunel
PY - 2026
DA - 2026/11/01
PB - arXiv
DO - 10.48550/arxiv.2611.01724
UR - https://arxiv.org/abs/2611.01724
AB - Conformal prediction (CP) provides a distribution-free framework for uncertainty quantification with finite-sample marginal coverage guarantees. Yet, missing values introduce distributional heterogeneity, where marginal coverage can hide systematic undercoverage for certain missingness patterns. At the same time, existing methods based on distances or smoothness require a continuous representation of partially observed covariates. Although imputation restores continuity, it obscures the missingness patterns and introduces task-dependent overhead. We develop a new CP framework that is \emph{adaptive to missingness patterns}, without imputing covariates. We rely on data augmentation by feature embedding that accounts for missingness. Our framework represents each input with missing values by a tree embedding together with its raw missingness mask, and then learns a regularized quantile score function. Varying the choices for the penalty function and the augmentation enables us to characterize three practically relevant validity levels: exact coordinate-wise validity, exact validity on observed masks, and approximate missingness-conditional guarantees for general mechanisms (MCAR, MAR, MNAR). Experiments on synthetic and real datasets support these guarantees and show that the proposed method reduces miscoverage risk while maintaining informative intervals, especially in non-MCAR settings. These results highlight the practical value of uncertainty quantification that adapts to missingness.
ER -