We analyse bill cosponsorship networks in the Italian Chamber of Deputies. Incomparison with other parliaments, a distinguishing feature of the Chamber isthe large number of political groups. Our analysis aims to infer the pattern ofcollaborations between these groups from data on bill cosponsorships. Wepropose an extension of stochastic block models for edge-valued graphs andderive measures of group productivity and of collaboration between politicalparties. As the model proposed encloses a large number of parameters, we pursuea penalized likelihood approach that enables us to infer a sparse reduced graphdisplaying collaborations between political parties.
展开▼