A0332
Title: The SIR model under missing data: A marginal likelihood approach via dynamical survival analysis
Authors: Suchismita Roy - Duke University (United States) [presenting]
Jason Xu - University of California Los Angeles (United States)
Alexander Fisher - Duke University (United States)
Abstract: The SIR model is widely used for modeling epidemic dynamics. Despite its widespread use, parameter inference in the presence of missing data is challenging due to the intractability of the likelihood in such compartmental models. To address this, a closed-form likelihood is developed for incidence data using the dynamical survival analysis (DSA) method, which provides a survival analysis-based interpretation of the SIR model. The method is flexible and computationally efficient. Through simulation, its performance is compared in parameter estimation with other methods that rely on the exact posterior of the SIR model. To further demonstrate the adaptability of the approach, the likelihood is extended to frailty models, illustrating how it can be modified to incorporate individual heterogeneity. Finally, the method is applied to real-world data, demonstrating its practical utility for epidemic inference with limited observations.
