Bayesian Methods in Complex Problems
Complex systems arise in many disciplines, from environmental sciences to engineering, from biology to medicine, just to name few fields. Models are proposed to describe their behaviour and their implementation requires the cooperation of statisticians, probabilists, numerical analysts and applied mathematicians in general, computer scientists and experts from the proper field, like engineers, biologists, medical doctors, economists, etc.
In the talk, the speaker will present some of the work he and his co-authors have done in different fields, namely predator-prey models in a context of integrated pest management, study of the thermal properties of a polymer (Plexiglas), degradation of cylinder liners in ships, evolution of Chlamydia infection and simulation of human circulation.
All those problems can be modelled using (partial) differential equations and introducing stochasticity taking care, e.g., of measurement errors, environmental and biological variability, and corrosive actions. The analysis of the systems implies, in all the cases, significant modelling, inferential, numerical and computational issues which will be illustrated in the talk.