Temporal Gaussian Copula For Clinical Multivariate Time Series Data Imputation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Su, Ye, Qiao, Hezhe, Wu, Di, Chen, Yuwen, Chen, Lin
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910902571237376
author Su, Ye
Qiao, Hezhe
Wu, Di
Chen, Yuwen
Chen, Lin
author_facet Su, Ye
Qiao, Hezhe
Wu, Di
Chen, Yuwen
Chen, Lin
contents The imputation of the Multivariate time series (MTS) is particularly challenging since the MTS typically contains irregular patterns of missing values due to various factors such as instrument failures, interference from irrelevant data, and privacy regulations. Existing statistical methods and deep learning methods have shown promising results in time series imputation. In this paper, we propose a Temporal Gaussian Copula Model (TGC) for three-order MTS imputation. The key idea is to leverage the Gaussian Copula to explore the cross-variable and temporal relationships based on the latent Gaussian representation. Subsequently, we employ an Expectation-Maximization (EM) algorithm to improve robustness in managing data with varying missing rates. Comprehensive experiments were conducted on three real-world MTS datasets. The results demonstrate that our TGC substantially outperforms the state-of-the-art imputation methods. Additionally, the TGC model exhibits stronger robustness to the varying missing ratios in the test dataset. Our code is available at https://github.com/MVL-Lab/TGC-MTS.
format Preprint
id arxiv_https___arxiv_org_abs_2504_02317
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Temporal Gaussian Copula For Clinical Multivariate Time Series Data Imputation
Su, Ye
Qiao, Hezhe
Wu, Di
Chen, Yuwen
Chen, Lin
Machine Learning
Artificial Intelligence
The imputation of the Multivariate time series (MTS) is particularly challenging since the MTS typically contains irregular patterns of missing values due to various factors such as instrument failures, interference from irrelevant data, and privacy regulations. Existing statistical methods and deep learning methods have shown promising results in time series imputation. In this paper, we propose a Temporal Gaussian Copula Model (TGC) for three-order MTS imputation. The key idea is to leverage the Gaussian Copula to explore the cross-variable and temporal relationships based on the latent Gaussian representation. Subsequently, we employ an Expectation-Maximization (EM) algorithm to improve robustness in managing data with varying missing rates. Comprehensive experiments were conducted on three real-world MTS datasets. The results demonstrate that our TGC substantially outperforms the state-of-the-art imputation methods. Additionally, the TGC model exhibits stronger robustness to the varying missing ratios in the test dataset. Our code is available at https://github.com/MVL-Lab/TGC-MTS.
title Temporal Gaussian Copula For Clinical Multivariate Time Series Data Imputation
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2504.02317