Time series models are widely used to examine temporal dynamics and uncover patterns across diverse fields. A commonly employed approach for modeling such data is the (Vector) Autoregressive (AR/VAR) model, in which each variable is represented as a linear combination of its own and others’ lagged values. However, the traditional (V)AR framework relies on the key assumption of stationarity, that autoregressive coefficients remain constant over time, which is often violated in practice, especially in systems affected by structural breaks, seasonal fluctuations, or evolving causal mechanisms. To overcome this limitation, Time-Varying (Vector) Autoregressive (TV-AR/TV-VAR) models have been developed, enabling model parameters to evolve over time and thus better capturing non-stationary behavior. Conventional approaches to estimating such models, including generalized additive modeling and kernel smoothing techniques, often require strong assumptions about basis functions, which can restrict their flexibility and applicability. To address these challenges, we introduce a novel framework that leverages physics-informed neural networks (PINN) to model TV-AR/TV-VAR processes. The proposed method extends the PINN framework to time series analysis by reducing reliance on explicitly defined physical structures, thereby broadening its applicability. Its effectiveness is validated through simulations on synthetic data and an empirical study of real-world health-related time series.
7 June 2021 in Health Information Science and Systems
This study examines the impact of the Affordable Care Act (ACA) on health insurance coverage among rent‐burdened households—those spending more than 30% of their income on rent—and non‐rent‐burdened households. Using data from American Community Survey, we find that Medicaid take‐up rate increased 8.88 percentage points (pp) among rent‐burdened households and 7.54 pp among non‐rent‐burdened households in expansion states. Conditional on household income and demographic characteristics, rent‐burdened households exhibit a 1.5 pp higher likelihood of Medicaid enrollment, with an additional decline of 0.7 pp in employer‐sponsored insurance and 1.0 pp in directly purchased insurance enrollment. These effects were more pronounced among individuals aged over 26 and those in states without state‐run exchanges. The findings show the importance of tailored Medicaid policies to assist households facing housing burdens, especially for those ineligible for housing vouchers.
Introduction: Housing is a critical social determinant of health, and individuals living in substandard housing often face greater healthcare needs and financial burdens. Expanding health insurance is therefore especially important for this population. However, little is known about whether the Affordable Care Act's Medicaid expansion affected disparities in coverage by housing quality. We examined whether Medicaid expansion improved insurance coverage equitably among low-income individuals living in homes with and without basic housing necessities. Methods: Using nationally representative survey data from 2010-2019, we compared changes in insurance coverage before and after expansion in states that expanded Medicaid eligibility vs those that did not. Results: Medicaid expansion increased Medicaid coverage by more than 12% points and reduced uninsurance by 6%-7% points relative to non-expansion states among individuals lacking at least one basic housing necessity. However, compared with individuals with complete housing necessities, those lacking at least one basic housing necessity experienced slightly smaller increases in Medicaid coverage and smaller reductions in uninsurance. Conclusion: Medicaid expansion improved coverage among individuals facing housing-related disadvantage but did not eliminate disparities. Eligibility-based expansions can extend coverage to structurally disadvantaged populations, but additional efforts to address non-financial barriers may be needed.