Information Network Modeling for U.S. Banking Systemic Risk |
| |
| Authors: | Giancarlo Nicola Paola Cerchiello Tomaso Aste |
| |
| Institution: | 1.Department of Economics and Management, University of Pavia, 27100 Pavia, Italy; (G.N.); (P.C.);2.Department of Computer Science, University College London, London WC1E 6EA, UK;3.Systemic Risk Centre, London School of Economics, London WC2A 2AE, UK |
| |
| Abstract: | In this work we investigate whether information theory measures like mutual information and transfer entropy, extracted from a bank network, Granger cause financial stress indexes like LIBOR-OIS (London Interbank Offered Rate-Overnight Index Swap) spread, STLFSI (St. Louis Fed Financial Stress Index) and USD/CHF (USA Dollar/Swiss Franc) exchange rate. The information theory measures are extracted from a Gaussian Graphical Model constructed from daily stock time series of the top 74 listed US banks. The graphical model is calculated with a recently developed algorithm (LoGo) which provides very fast inference model that allows us to update the graphical model each market day. We therefore can generate daily time series of mutual information and transfer entropy for each bank of the network. The Granger causality between the bank related measures and the financial stress indexes is investigated with both standard Granger-causality and Partial Granger-causality conditioned on control measures representative of the general economy conditions. |
| |
| Keywords: | granger causality graphical models financial stress |
|
|
