CMStatistics 2018: Start Registration
View Submission - CMStatistics
B0930
Title: Heterogeneous susceptibilities in network influence models Authors:  Daniel Sewell - University of Iowa (United States) [presenting]
Abstract: Network autocorrelation models are widely used to evaluate the impact of influence on some variable of interest or diffusion through a network. This is a large class of models that parsimoniously accounts for how one's neighbors influence one's own behaviors, opinions, or states by incorporating the network adjacency matrix into the joint distribution of the data. These models assume homogeneous susceptibility to influence through the network, however, which may be a strong assumption in many contexts. A hierarchical model is proposed which allows the influence parameter to be a function of individual attributes and/or of local network topological features. An approximation of the posterior distribution is derived in a general framework that is applicable to the Durbin, network effects, network disturbances, or network moving average autocorrelation models. The proposed approach can also be applied to investigating determinants of influence in the context of egocentric network data.