Flaminia Musella

Background

About Flaminia Musella

Research interests. Her research interests are in the area of multivariate statistical techniques both from a methodological-computational and applicative point of view. She is interested in computational and applicative aspects of graphical models and, in particular, of Bayesian networks, for the analysis of the dependencies between ordinal and mixed nominal-ordinal variables. In her doctoral thesis, she worked on a new algorithm for learning the independence structures between categorical variables. Applications: 1) Education. She is focusing on the effects of distance learning for educational outcomes by accounting for emotional impacts on students. She deals with multivariate statistical methods for the analysis of the diffusion of digital skills for teachers of schools and University. 2) Gender. Since recently, she is occupying of gender data for evaluating social and economics gender gap. She is involved in research project with Roma Tre University, Bicocca University and Poligrafico dello Stato for highlighting gender indicators useful for producing the gender report and the gender action plan. Similarly, she is involved in some conseulling activities for private companies for supporting them in the gender reporting. 3) Socio-Economic. She has worked on Bayesian network models for the analysis of energy markets. She dealt with multivariate statistical methods for the study of economic-environmental phenomena starting from complex data. In particular, she focused on the study of conflicts between stakeholders for the control and management of resources in fragile areas such as coastal areas. 4) Managerial application. She is interested in the business applications of graphic models and, in particular, of Bayesian networks with a particular focus on the quality measurement and monitoring of public services provided on-line or off-line She has developed interest in the application of Bayesian networks for decisions as a tool to support business decisions and as an information management tool for improving the quality of services. She works on the quantitative management of complex business problems integrating different quality aspects through Bayesian Object-Oriented networks. 5) Computational aspects. She dealt with constraint-based algorithms for non-normal data with the help of copula models. She proposed a new strategies for learning dependencies structures among categorical data. During her stay at the University of Oxford she collaborated on the creation of the gRim package with David Edwards and Søren Højsgaard.