Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Can the Christoffersen Test Strengthen Financial Stability?

Zukauskas, Tadas LU (2025) NEKN02 20251
Department of Economics
Abstract (Swedish)
This thesis investigates whether incorporating the Christoffersen independence
test into regulatory backtesting frameworks can enhance financial stability by improving the detection of systemic risk. Using daily Value-at-Risk and profit-and-loss data from 13 major international banks over 2018–2022, the study distinguishes between clustered and isolated VaR exceptions, applying event studies and panel
regressions with market and bank-level controls. The results show that while both
clustered and isolated exceptions are associated with increased abnormal volatility, only isolated events remain statistically significant for bank-level volatility after accounting for time effects. However, clustered exceptions have a substantial and robust... (More)
This thesis investigates whether incorporating the Christoffersen independence
test into regulatory backtesting frameworks can enhance financial stability by improving the detection of systemic risk. Using daily Value-at-Risk and profit-and-loss data from 13 major international banks over 2018–2022, the study distinguishes between clustered and isolated VaR exceptions, applying event studies and panel
regressions with market and bank-level controls. The results show that while both
clustered and isolated exceptions are associated with increased abnormal volatility, only isolated events remain statistically significant for bank-level volatility after accounting for time effects. However, clustered exceptions have a substantial and robust impact on market-wide volatility (VIX), and Granger causality tests reveal that clusters precede spikes in systemic risk. The evidence suggests that the Christoffersen test provides valuable warning signals for market instability, which the current Basel traffic light approach may overlook. The findings support policy recommendations to supplement or revise regulatory backtesting regimes by incorporating independence testing, thereby strengthening the resilience of the financial system during periods of stress. (Less)
Please use this url to cite or link to this publication:
author
Zukauskas, Tadas LU
supervisor
organization
course
NEKN02 20251
year
type
H1 - Master's Degree (One Year)
subject
keywords
Value-at-Risk, Christoffersen Test, Financial Stability, Exception Clustering, Risk Regulation
language
English
id
9193571
date added to LUP
2025-09-12 10:47:19
date last changed
2025-09-12 10:47:19
@misc{9193571,
  abstract     = {{This thesis investigates whether incorporating the Christoffersen independence
test into regulatory backtesting frameworks can enhance financial stability by improving the detection of systemic risk. Using daily Value-at-Risk and profit-and-loss data from 13 major international banks over 2018–2022, the study distinguishes between clustered and isolated VaR exceptions, applying event studies and panel
regressions with market and bank-level controls. The results show that while both
clustered and isolated exceptions are associated with increased abnormal volatility, only isolated events remain statistically significant for bank-level volatility after accounting for time effects. However, clustered exceptions have a substantial and robust impact on market-wide volatility (VIX), and Granger causality tests reveal that clusters precede spikes in systemic risk. The evidence suggests that the Christoffersen test provides valuable warning signals for market instability, which the current Basel traffic light approach may overlook. The findings support policy recommendations to supplement or revise regulatory backtesting regimes by incorporating independence testing, thereby strengthening the resilience of the financial system during periods of stress.}},
  author       = {{Zukauskas, Tadas}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Can the Christoffersen Test Strengthen Financial Stability?}},
  year         = {{2025}},
}