An MVAR Framework to Capture Extreme Events in Macro-Prudential Stress Tests

46 Pages Posted: 5 Sep 2012

See all articles by Paolo Guarda

Paolo Guarda

Banque Centrale du Luxembourg

Abdelaziz Rouabah

Banque Centrale du Luxembourg

John Theal

Banque Centrale du Luxembourg

Date Written: August 13, 2012

Abstract

Severe financial turbulences are driven by high impact and low probability events that are the characteristic hallmarks of systemic financial stress. These unlikely adverse events arise from the extreme tail of a probability distribution and are therefore very poorly captured by traditional econometric models that rely on the assumption of normality. In order to address the problem of extreme tail events, we adopt a mixture vector autoregressive (MVAR) model framework that allows for a multi-modal distribution of the residuals. A comparison between the respective results of a VAR and MVAR approach suggests that the mixture of distributions allows for a better assessment of the effect that adverse shocks have on counterparty credit risk, the real economy and banks’ capital requirements. Consequently, we argue that the MVAR provides a more accurate assessment of risk since it captures the fat tail events often observed in time series of default probabilities.

Keywords: Stress testing, MVAR, tier 1 capital ratio, counterparty risk, Luxembourg banking sector

JEL Classification: C15, E44, G01, G21

Suggested Citation

Guarda, Paolo and Rouabah, Abdelaziz and Theal, John, An MVAR Framework to Capture Extreme Events in Macro-Prudential Stress Tests (August 13, 2012). ECB Working Paper No. 1464. Available at SSRN: https://ssrn.com/abstract=2128483

Paolo Guarda (Contact Author)

Banque Centrale du Luxembourg ( email )

2, boulevard Royal
Luxembourg, L-2983
Luxembourg

Abdelaziz Rouabah

Banque Centrale du Luxembourg ( email )

2, boulevard Royal
Luxembourg, L-2983
Luxembourg

John Theal

Banque Centrale du Luxembourg ( email )

2 Boulevard Royal
Luxemburg, L-2983
Luxembourg

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
57
Abstract Views
446
rank
382,974
PlumX Metrics