Template-Type: ReDIF-Paper 1.0 Series: Tinbergen Institute Discussion Papers Creation-Date: 2017-10-26 Number: 17-101/II Author-Name: Fernando Linardi Author-Email: fernando.linardi@bcb.gov.br Author-Workplace-Name: University of Amsterdam, The Netherlands; Central Bank of Brazil, Brazil Author-Name: Cees (C.G.H.) Diks Author-Email: C.G.H.Diks@uva.nl Author-Workplace-Name: University of Amsterdam, The Netherlands; Tinbergen Institute, The Netherlands Author-Name: Marco (M.J.) van der Leij Author-Email: M.J.vanderLeij@uva.nl Author-Workplace-Name: University of Amsterdam, The Netherlands; De Nederlandsche Bank, The Netherlands Author-Name: Iuri Lazier Author-Email: iuri.lazier@bcb.gov.br Author-Workplace-Name: Central Bank of Brazil, Brazil Title: Dynamic Interbank Network Analysis Using Latent Space Models Abstract: Longitudinal network data are increasingly available, allowing researchers to model how networks evolve over time and to make inference on their dependence structure. In this paper, a dynamic latent space approach is used to model directed networks of monthly interbank exposures. In this model, each node has an unobserved temporal trajectory in a low-dimensional Euclidean space. Model parameters and latent banks' positions are estimated within a Bayesian framework. We apply this methodology to analyze two different datasets: the unsecured and the secured (repo) interbank lending networks. We show that the model that incorporates a latent space performs much better than the model in which the probability of a tie depends only on observed characteristics; the latent space model is able to capture some features of the dyadic data such as transitivity that the model without a latent space is not able to. Classification-JEL: C11, D85, G21 Keywords: network dynamics, latent position model, interbank network, Bayesian inference File-Url: https://papers.tinbergen.nl/17101.pdf File-Format: application/pdf File-Size: 885693 bytes Handle: RePEc:tin:wpaper:20170101