Pierre Siklos
- 19 July 2018
- WORKING PAPER SERIES - No. 78Details
- Abstract
- We use a factor model and elastic net shrinkage to model a high-dimensional network of European CDS spreads. Our empirical approach allows us to assess the joint transmission of bank and sovereign risk to the non-financial corporate sector. Our findings identify a sectoral clustering in the CDS network, where financial institutions are in the center and non-financial entities as well as sovereigns are grouped around the financial center. The network has a geographical component reflected in different patterns of real-sector risk transmission across countries. Our framework also provides dynamic estimates of risk transmission, a useful tool for systemic risk monitoring.
- JEL Code
- C32 : Mathematical and Quantitative Methods→Multiple or Simultaneous Equation Models, Multiple Variables→Time-Series Models, Dynamic Quantile Regressions, Dynamic Treatment Effect Models, Diffusion Processes
C38 : Mathematical and Quantitative Methods→Multiple or Simultaneous Equation Models, Multiple Variables→Classification Methods, Cluster Analysis, Principal Components, Factor Models
C55 : Mathematical and Quantitative Methods→Econometric Modeling→Modeling with Large Data Sets?
F3 : International Economics→International Finance
G01 : Financial Economics→General→Financial Crises
G15 : Financial Economics→General Financial Markets→International Financial Markets