A goodness of fit approach to select the optimal visit sequence for nonparametric data synthesis
Abstract
Synthetic data are a promising way to maintain privacy while disseminating research data, thereby enabling compliance with the open science principles. Popular methods for generating synthetic data include sequential synthesis, where each variable is generated conditionally on the preceding ones in a given order. This order, also called the visit sequence of the synthesizer, is known to affect the quality of synthetic data. This raises the question of the criteria that should guide its choice. In this work, we present a novel method to perform such selection based on a goodness of fit criterion, and show that it outperforms alternative structure-based methods, both in terms of synthesis quality and time complexity. The proposed method also has the advantage of being applicable to any type of data and any class of conditional distribution models.
Related
- Optimal estimation and goodness-of-fit testing of the mean for sparse longitudinal functional data
- An Optimal Bound for Ramsey Goodness of Cycles
- Optimal Neural Estimation for High-Dimensional Nonparametric Regression under Multisource Covariate Shift
- Optimal Nonparametric Estimation of Phase-Space Representations for Non-Gaussian Continuous Variable Quantum States
- Optimal Synthesis on the Special Linear Group
BibTeX
@misc{chapelle2026goodness,
title = {{A goodness of fit approach to select the optimal visit sequence for nonparametric data synthesis}},
author = {Rémy Chapelle and Bruno Falissard and Mohammed Sedki},
howpublished = {HAL},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.38559},
eprint = {2610.38559},
archivePrefix = {arXiv},
url = {https://hal.science/hal-05835785v1},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - A goodness of fit approach to select the optimal visit sequence for nonparametric data synthesis
AU - Rémy Chapelle
AU - Bruno Falissard
AU - Mohammed Sedki
PY - 2026
DA - 2026/10/27
PB - HAL
DO - 10.48550/arxiv.2610.38559
UR - https://hal.science/hal-05835785v1
AB - Synthetic data are a promising way to maintain privacy while disseminating research data, thereby enabling compliance with the open science principles. Popular methods for generating synthetic data include sequential synthesis, where each variable is generated conditionally on the preceding ones in a given order. This order, also called the visit sequence of the synthesizer, is known to affect the quality of synthetic data. This raises the question of the criteria that should guide its choice. In this work, we present a novel method to perform such selection based on a goodness of fit criterion, and show that it outperforms alternative structure-based methods, both in terms of synthesis quality and time complexity. The proposed method also has the advantage of being applicable to any type of data and any class of conditional distribution models.
ER -