Current students - PhD DADS

RAGNI ALESSANDRA

Cycle: XXXVII

Advisor: PAGANONI ANNA MARIA
Tutor: IEVA FRANCESCA

Major Research topic:
Generalizing mixed effects models for modelling complex hierarchical data

Abstract:

During my first year of PhD, coming in touch with administrative datasets in the fields of learning analytics and healthcare, I realized how frequently hierarchical data arise. The hierarchy is typically given by students in schools or patients in clinical centres. The hierarchical structure of such data must be taken into account in the modelling procedure, motivating the use of specific models tailored for hierarchical data such as mixed effect models. The more specific object of my research is the development of novel generalizations of mixed effects models with the most various outcomes (i.e., Bernoulli, Poisson or time to event data) able to deal with complex hierarchical data.
Semi-parametric assumptions on the random effects distribution in the generalized linear models will be explored for the development of new methodologies and the statistical analysis and prediction of quantities of interest. In particular, generalized mixed effects models producing a clustering within the highest level of the hierarchy is being developed for binary and integer responses. Particular attention will be devoted to the effectiveness, performance and applicability of the proposed methods.