Sophisticated vs. Simple Systemic Risk Measures
Series
WORKING PAPERS ON FINANCE
Type
working paper
Date Issued
2014
Author(s)
Abstract
This paper evaluates whether sophisticated or simple systemic risk measures are more suitable in identifying which institutions contribute to systemic risk. In this investigation, DCoVaR, Marginal Expected Shortfall (MES), SRISK and Granger-Causality Networks are considered as sophisticated systemic risk measures. Market capitalization, total debt, leverage, the stock market returns of an institution, and the correlation between the stock market returns of an institution and the market, are considered as simple systemic risk measures. Systemic relevance is approximated by the receipt of financial support during the financial crisis and the classification, as a systemically important institution, by national or international regulators. The analyses are performed for all companies included in the S&P 500 composite index. The findings suggest that simple systemic risk measures have more explanatory power than sophisticated risk measures. In particular, total debt is found to be the most suitable indicator to detect institutions which contribute to systemic risk, according to the explanatory power and model fit. The most suitable sophisticated risk measure seems to be SRISK.
Language
English
Keywords
Systemic Risk
DCoVaR
Marginal Expected Shortfall
SRISK
Granger-Causality Networks
HSG Classification
contribution to scientific community
Refereed
No
Number
2014/22
Subject(s)
Division(s)
Eprints ID
238470
File(s)![Thumbnail Image]()
Loading...
open.access
Name
14_22_Pankoke_Sophisticated vs Simple Systemic Risk Mearurement.pdf
Size
336.15 KB
Format
Adobe PDF
Checksum (MD5)
b4620c0bf27cc5627f5d14d31c26977f