Graduate Students Seminar

Speaker 1: Nadeesri Wijekoon

Title

Statistical Modeling Using Conditionally Specified Joint Distributions

Abstract

Often in practice, conditionals are easier to model and interpret while the joint distribution itself is either intractable or not available in closed form. When the observed response consists of both continuous and discrete components, specifying conditionals is more convenient; and knowing the conditional distributions makes it easier to understand and visualize the joint distribution. Such joint distributions are referred to as conditionally specified models.

In this presentation, we will provide several practical motivating examples where this approach is intuitively appealing and introduce a theorem which can use to derive conditionally specified models and discuss the theoretical aspects of these models.

Speaker 2: Maria Deliyianni

Title

Damping in elastic systems

Abstract

An inextensible cantilever undergoing large deflections is under consideration. In order to answer questions about well posedness of the model, we need to modify the equations of motion by adding an additional term, the so-called Kelvin-Voigt damping. Damping describes the energy dissipation in elastic systems resulting from a variety of sources, both internal and external. Weak damping, strong and square root are some of the mathematical forms that are used to capture the phenomenon of energy leaving the system. In this talk, we examine each of them in the context of the Euler-Bernoulli beam with hinged boundary conditions.  Furthermore, we discuss the possibility of weakening the strength of the damping we have added in the aforementioned model and explain why, even though square root damping is a good candidate for improving the result, it is not physically meaningful for a cantilever.