Bayesian modelling of temporal bipartite networks using a latent space approach with applications to interlocking directorates in Irish companies
This talk is concerned with the development of Bayesian models and inferential techniques to handle the situation of a bipartite network which evolves in time. Data of this type occurs in many situations in biology and sociology, for example. In this talk we analyze the temporal bipartite network of the leading Irish companies and their directors from 2003 to 2013, encompassing the end of the Celtic Tiger boom and the ensuing financial crisis in 2008. We focus on the evolution of company interlocks, whereby a company director simultaneously sits on two or more boards. We develop a statistical model for this dataset by embedding the positions of companies and directors in a latent space. The temporal evolution of the network is modeled through three levels of Markovian dependence: one on the model parameters, one on the companies’ latent positions, and one on the edges themselves. The model is estimated using Bayesian inference. Our analysis reveals that the level of interlocking, as measured by a contraction of the latent space, increased before and during the crisis, reaching a peak in 2009, and has generally stabilized since then.
This work is joint with Riccardo Rastelli, Jason Wyse and Adrian Raftery. Click here to find out more.
Professor Nial Friel: short bio
I am a professor of statistics in the school of mathematics and statistics, University College Dublin. I graduated with a PhD in statistics from the University of Glasgow in 1999. Following post-doctoral positions at Queensland University of Technology, University of Cambridge and Athens University of Economics and Business, I joined the University of Glasgow as a lecturer in 2002 and then as a reader from 2006. In 2007 I was appointed as an associate professor at University College Dublin and was subsequently appointed full professor in 2014. I hold an adjunct professorship from Queensland University of Technology since 2013. I am a co-PI in Insight, the national centre for data analysis where I lead the statistics and machine learning programme. My research interests lie broadly in Bayesian statistics, Monte Carlo methods and applications, particularly in statistical network analysis. View Professor Friel’s website.