Robust Spatiotemporal Epidemic Modeling with Integrated Adaptive Outlier Detection

2025-07-12

AI Summary

In epidemiological modeling, outliers often bias parameter estimation, compromising public health decision-making. To address this, we propose a Robust Spatio-Temporal Generalized Additive Model (RST-GAM) that jointly models disease transmission dynamics and detects outliers. Methodologically, RST-GAM incorporates mean-shift parameters and adaptive Lasso regularization to sparsely identify and correct anomalous observations; integrates triangulation-based penalized splines with data-pruning weighting to enhance spatio-temporal smoothness and robustness; and optimizes the convex quasi-negative log-likelihood via univariate/bivariate splines and an adaptive-step proximal algorithm, ensuring global convergence. We theoretically establish parameter estimation error bounds and variable selection consistency. Experiments demonstrate RST-GAM’s robustness under multiple outlier scenarios. Applied to U.S. county-level COVID-19 data, it significantly improves parameter estimation accuracy and hotspot detection capability.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Reasoning under Uncertainty: Stochastic Optimization
• Intelligent Robots: State Estimation

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

In epidemic modeling, outliers can distort parameter estimation and ultimately lead to misguided public health decisions. Although there are existing robust methods that can mitigate this distortion, the ability to simultaneously detect outliers is equally vital for identifying potential disease hotspots. In this work, we introduce a robust spatiotemporal generalized additive model (RST-GAM) to address this need. We accomplish this with a mean-shift parameter to quantify and adjust for the effects of outliers and rely on adaptive Lasso regularization to model the sparsity of outlying observations. We use univariate polynomial splines and bivariate penalized splines over triangulations to estimate the functional forms and a data-thinning approach for data-adaptive weight construction. We derive a scalable proximal algorithm to estimate model parameters by minimizing a convex negative log-quasi-likelihood function. Our algorithm uses adaptive step-sizes to ensure global convergence of the resulting iterate sequence. We establish error bounds and selection consistency for the estimated parameters and demonstrate our model's effectiveness through numerical studies under various outlier scenarios. Finally, we demonstrate the practical utility of RST-GAM by analyzing county-level COVID-19 infection data in the United States, highlighting its potential to inform public health decision-making.

Problem

• Detect and adjust outliers in epidemic modeling
• Model spatiotemporal disease spread robustly
• Improve public health decisions with accurate data

Innovation

• Robust spatiotemporal generalized additive model
• Adaptive Lasso regularization for outliers
• Scalable proximal algorithm with step-sizes