Saved in:
Bibliographic Details
Main Authors: Ounissi, Oussama, Jävergård, Nicklas, Muntean, Adrian
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.02405
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916985835618304
author Ounissi, Oussama
Jävergård, Nicklas
Muntean, Adrian
author_facet Ounissi, Oussama
Jävergård, Nicklas
Muntean, Adrian
contents This work introduces the application of the Orthogonal Procrustes problem to the generation of synthetic data. The proposed methodology ensures that the resulting synthetic data preserves important statistical relationships among features, specifically the Pearson correlation. An empirical illustration using a large, real-world, tabular dataset of energy consumption demonstrates the effectiveness of the approach and highlights its potential for application in practical synthetic data generation. Our approach is not meant to replace existing generative models, but rather as a lightweight post-processing step that enforces exact Pearson correlation to an already generated synthetic dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2510_02405
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Orthogonal Procrustes problem preserves correlations in synthetic data
Ounissi, Oussama
Jävergård, Nicklas
Muntean, Adrian
Methodology
Statistics Theory
Machine Learning
47A55, 15A18, 15-03
This work introduces the application of the Orthogonal Procrustes problem to the generation of synthetic data. The proposed methodology ensures that the resulting synthetic data preserves important statistical relationships among features, specifically the Pearson correlation. An empirical illustration using a large, real-world, tabular dataset of energy consumption demonstrates the effectiveness of the approach and highlights its potential for application in practical synthetic data generation. Our approach is not meant to replace existing generative models, but rather as a lightweight post-processing step that enforces exact Pearson correlation to an already generated synthetic dataset.
title Orthogonal Procrustes problem preserves correlations in synthetic data
topic Methodology
Statistics Theory
Machine Learning
47A55, 15A18, 15-03
url https://arxiv.org/abs/2510.02405