Predicting profitability using advice branch bank networks

The literature on social networks and their analysis has undergone explosive growth in the past decade. Network models have been used to study structures as diverse as the interaction of monks in a monastery, the links across the World Wide Web, and the structure of organizations. In much of this literature the network itself is viewed as the object of interest, and models are used to elucidate its structure. In this paper, we adopt a different perspective and we explore the role of network structure of organizations for prediction purposes. In particular, we work with data gathered on the advice-seeking habits of employees in 52 branches of a major North American bank corporation. We then use the network structure within each branch discovered via various exploratory analyses to predict the profitability of the individual branches.