The Dynamic Factor Network Model with an Application to Global Credit-Risk

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Abstract We introduce a dynamic network model with probabilistic link functions that depend on stochastically time-varying parameters. We adopt the widely used blockmodel framework and allow the high-dimensional vector of link probabilities to be a function of a low-dimensional set of dynamic factors. The resulting dynamic factor network model is straightforward and transparent by nature. However, parameter estimation, signal extraction of the dynamic factors, and the econometric analysis generally are intricate tasks for which simulation-based methods are needed. We provide feasible and practical solutions to these challenging tasks, based on a computationally e cient importance sampling procedure to evaluate the likelihood function. A Monte Carlo study is carried out to provide evidence of how well the methods work. In an empirical study, we use the novel framework to analyze a database of significance-flags of Granger causality tests for pair-wise credit default swap spreads of 61 di erent banks from the United States and Europe. Based on our model, we recover two groups that we characterize as "local" and "international" banks. The credit-risk spillovers take place between banks, from the same and from di erent groups, but the intensities change over time as we have witnessed during the financial crisis and the sovereign debt crisis. Terms of use: Documents in

A dynamic network model with persistent links and node-specific latent variables, with an application to the interbank market

“…They describe the link dynamics with a Tobit model, allowing for deterministic or stochastic time-varying parameters that take into account the possibility of structural changes in network dynamics. Brauning and Koopman [17] have applied the dynamic factors model to the case of dynamic networks. Depending on the number of factors, the model allows to reduce the dimensionality of the problem and to describe cross-sectional dependencies in network data.…”

“…the tendency of a link that does (or does not) exist at time t − 1 to continue existing (or not existing) at time t, similarly to [5]. From the other side, we describe the stochastic dynamics of node-specific latent variables that we call fitnesses, with a similar aim of [17,18]. The node fitness describes the tendency of a node in creating links and its evolution determines how the degree of the node changes in time.…”

Quantifying Systemic Risk in the Presence of Unlisted Banks: Application to the European Banking Sector

“…Wang (2021) suggest a deep learning approach. Alternatively, network models could be used to mimic the often observed core-periphery structure of the financial sector (Bräuning and Koopman, 2016;Andrieş et al, 2022).…”

“…Billio et al (2012) offer an early econometric model which quantifies interconnectedness through principal components and implied networks based on Granger-causality tests. Bräuning and Koopman (2016) extend the idea with time-varying heterogeneity in the link formation between banks using CDS spreads of US and European institutions, thus aiming to capture the dynamic formation of potential core-periphery clusters, which are natural for the financial sector. Moratis and Sakellaris (2021) on the other hand use a panel VAR model to decompose the transmission of systemic shocks across a universe of global banks.…”