Detalles de Evento
Este evento sucederá desde el 14 enero 2025 hasta el 15 enero 2025. Ocurrirá nuevamente el 14 enero 2025 09:00.
An overview on some stochastic models of growth
Speaker: Giuseppina Albano (Department of Economics and Statistics, University of Salerno, Italia).
Seminar 1:
Title: Growths equations and their stochastic generalizations: modeling and inference.
Abstract: In this seminar the introduction of stochasticity in growth equations is developed. In particular, the introduction of noise within the growth equations is discussed, distinguishing between multiplicative and additive noises. The resulting stochastic diffusion processes and their parametric inference procedures are discussed. Results for a general growth curve including a wide family of growth phenomena are shown. Maximum likelihood method is discussed, and its solution is obtained by means of metaheuristic techniques. Several simulation studies and an application to real data are performed. The discussion of the topics will be approached from an informal and not analytic point of view, to provide students with the basic idea of these topics, favoring the intuitive aspect over methodological rigor.
Date: 14th January 2025, at 9:00h.
Seminar 2:
Title: Stochastic SIR model including growths: modeling and inference.
Abstract: In this talk the focus is on epidemiological models. In particular, the background of Susceptible-Infected- Removed model is discussed, emphasizing the role of the contribution of each subpopulation. The role of stochasticity in such models is also addressed. However, the main contribution is made to the inference for SIR models. For a SIR stochastic model, able to include the natural growth of the Susceptible population, the inference is addressed by means of a quasi-maximum likelihood method. Numerical procedures for determining the local maxima of the likelihood function are discussed. Emphasis is placed on the problems associated with the use of such techniques and some insights are provided.
Date: 15th January 2025, at 9:00h.