Unsupervised Classification through Quantile Regression

This paper aims to propose an innovative approach to identify group effects through a quantile regression model. Quantile regression is a quite recent regression technique that allows to focus on the effects that a set of explanatory variables has on the entire conditional distribution of a dependent variable. The proposal concerns the use of multivariate techniques to detect effects attributable to different group membership and it is illustrated using an empirical analysis. In particular the impact of student features on the University outcome measured by the degree mark is evaluated taking into account that the dependence structure could be different according to the Faculty membership.