Mixed Engression: Distributional Regression for Multivariate Mixed-Type Responses
Abstract
Estimating the conditional distribution of multivariate responses with continuous and binary components requires capturing their dependence, respecting their different supports, and accommodating potentially heavy-tailed continuous outcomes. We propose mixed engression, which fits a continuous generator using energy losses induced by bounded discrepancies and then thresholds its soft binary outputs to produce mixed-response samples. Because the discrepancies are bounded, the population objective remains well defined without response-moment assumptions. However, we show that the full-response energy distance for the complete response vector can tend to zero before thresholding yet remain bounded away from zero afterward as response dimensions grow. A separate energy loss on the entire binary vector captures discrepancies obscured by the full-response comparison and controls the effect of thresholding. The generator is therefore fitted using an augmented energy loss that combines the full-response and binary components. Under compositional smoothness of the structural functions and a uniform threshold-margin condition, we derive nonasymptotic bounds on the final sampler's full-response error for sparse ReLU estimators of models driven by Gaussian or Student-t structural noise, yielding polynomial convergence rates up to logarithmic factors. We also establish consistency when the covariate, response, and noise dimensions grow under suitable conditions. Simulations show improved accuracy relative to the comparison methods under the full-response energy distance, including in heavy-tailed settings. Moderate binary augmentation substantially improves binary-distribution accuracy while largely preserving full-response accuracy.
Related
- Belted Engression: Sufficient Dimension Reduction for Generative Distributional Regression
- A Dirichlet Mixed-Membership Model for Exact Multivariate Distributional Credibility
- On the Calibration of Isotonic Distributional Regression
- Distributional Regression for Arbitrary Response Variables Using ReLU Neural Networks
- Sufficiently Reduced Distributional Regression
BibTeX
@misc{hwang2026mixed,
title = {{Mixed Engression: Distributional Regression for Multivariate Mixed-Type Responses}},
author = {Seyong Hwang and Kean Ming Tan and Ji Zhu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.32850},
eprint = {2610.32850},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.32850},
}
RIS
TY - UNPB
TI - Mixed Engression: Distributional Regression for Multivariate Mixed-Type Responses
AU - Seyong Hwang
AU - Kean Ming Tan
AU - Ji Zhu
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2610.32850
UR - https://arxiv.org/abs/2610.32850
AB - Estimating the conditional distribution of multivariate responses with continuous and binary components requires capturing their dependence, respecting their different supports, and accommodating potentially heavy-tailed continuous outcomes. We propose mixed engression, which fits a continuous generator using energy losses induced by bounded discrepancies and then thresholds its soft binary outputs to produce mixed-response samples. Because the discrepancies are bounded, the population objective remains well defined without response-moment assumptions. However, we show that the full-response energy distance for the complete response vector can tend to zero before thresholding yet remain bounded away from zero afterward as response dimensions grow. A separate energy loss on the entire binary vector captures discrepancies obscured by the full-response comparison and controls the effect of thresholding. The generator is therefore fitted using an augmented energy loss that combines the full-response and binary components. Under compositional smoothness of the structural functions and a uniform threshold-margin condition, we derive nonasymptotic bounds on the final sampler's full-response error for sparse ReLU estimators of models driven by Gaussian or Student-t structural noise, yielding polynomial convergence rates up to logarithmic factors. We also establish consistency when the covariate, response, and noise dimensions grow under suitable conditions. Simulations show improved accuracy relative to the comparison methods under the full-response energy distance, including in heavy-tailed settings. Moderate binary augmentation substantially improves binary-distribution accuracy while largely preserving full-response accuracy.
ER -