Temporal Gaussian Copula For Clinical Multivariate Time Series Data Imputation
This work addresses the challenge of data imputation for clinical multivariate time series, which is crucial for improving data quality in healthcare applications, but it appears incremental as it builds on existing statistical and deep learning methods with a novel hybrid approach.
The paper tackles the problem of imputing missing values in multivariate time series data, which often has irregular missing patterns, by proposing a Temporal Gaussian Copula Model (TGC) that leverages Gaussian Copula to capture cross-variable and temporal relationships, and it demonstrates substantial outperformance over state-of-the-art methods in experiments on three real-world datasets.
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.