This item is available under a Creative Commons License for non-commercial use only
In this paper we introduce a new modelling approach to analyse weighted signed networks by assuming that their generative process consists of two models: the interaction model which describes the overall connectivity structure of the relations in the network without taking into account neither the weight nor the sign of the dyadic relations; and the conditional weighted signed network model describes how the edge signed weights form given the interaction structure. We then show how this modelling approach can facilitate the interpretation of the overall network process. Finally, we adopt a Bayesian inferential approach to illustrate the new methodology by modelling the Sampson’s influence network.
Caimo, A. & Gollini, I. (2019), Modelling Weighted Signed Networks. Proceedings of the Conference of the Italian Statistical Society pg. 111 – 118. doi:10.21427/mear-8r46