Answers are grounded in this paper and cite their sources.
This paper develops the mathematical theory of multivariate multi-sigmoidal Gompertz and Weibull diffusion processes for modeling complex growth phenomena with multiple inflection points. We extend univariate formulations to a multivariate framework by incorporating time-dependent and fixed covariates through a flexible polynomial structure Q(t,z) of degree four, enabling the representation of up to three distinct growth phases. The deterministic ordinary differential equations are derived, closed-form solutions are obtained, and stochastic extensions via Itô calculus are rigorously established. We derive the complete maximum likelihood estimation framework, including the score function, observed Hessian, and Fisher information matrix. The theoretical results demonstrate that a fourth-degree polynomial yields up to three inflection points, consistent with multi-sigmoidal behavior and enabling the representation of three distinct growth phases. The models are applied to US COVID-19 data as a demonstration of the mathematical framework, with parameter estimates confirming the theoretical properties and revealing the practical limitations of the lognormal assumption. This work provides a comprehensive mathematical foundation for multivariate multi-sigmoidal diffusion processes with applications in epidemiology, biology, and economics.
No takes yet. Share an insight, caveat, or question.
Oyamakin et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: