Sovereign Risk: How Filtered Bootstrap and Historical Simulation Catch Government Problems

28 Pages Posted: 6 Dec 2011 Last revised: 8 Dec 2011

Stefano Colucci

Symphonia Sgr; University of Rome III - Department of Business Studies

Dario Brandolini

University of Turin

Date Written: November 4, 2011

Abstract

The purpose of this paper is to apply the results of Brandolini D. – Colucci S. “Backtesting Value-at-Risk: A comparison between Filtered Bootstrap and Historical Simulation” in order to extend the VaR estimation also in a bond universe and particularly in order to estimate sovereign risk. We see a comparison between two risk models to evaluate, VaR, Historical Simulation and Monte Carlo Filtered Bootstrap. We perform three tests, Unconditional Coverage, Independence and Conditional Coverage as in Christoffersen, P., Pellettier D. (2004) paper. We present results on both VaR 1% both VaR 5% in one day horizon for the two models on seven sovereign bond proxed by the following indices: Merril Lynch German Federal Governments, 7-10y (G4D0 as BUND), Merril Lynch Portuguese Governments, 7-10y (G4U0 as PGB), Merril Lynch Irish Governments, 7-10y (G4R0 as IRISH), Merril Lynch Italian Governments, 7-10y (G4I0 as BTP), Merril Lynch Spanish Governments, 7-10y (G4E0 as BONOS), Merril Lynch Emerging Market Sovereigns, Greece (GDGR as GGB) and Merril Lynch US Treasury 7-10y (G4A0) all in Local Currency. Our results show that Filtered Bootstrap Approach satisfy all test for all indices, while Historical Simulation has many rejection cases. Finally we also test in a regulatory framework (rolling window of 250 observations) the two models and the advantages of using a conditional coverage methodology to validate risk models.

Keywords: VaR, Backtest, Filtered Bootstrap, Historical Simalation, Sovereign Risk

JEL Classification: C22, G22, G23, G24

Suggested Citation

Colucci, Stefano and Brandolini, Dario, Sovereign Risk: How Filtered Bootstrap and Historical Simulation Catch Government Problems (November 4, 2011). Available at SSRN: https://ssrn.com/abstract=1968117 or http://dx.doi.org/10.2139/ssrn.1968117

Stefano Colucci (Contact Author)

Symphonia Sgr ( email )

via Gramsci 7
Torino, Torino 10144
Italy

University of Rome III - Department of Business Studies ( email )

Via Silvio D'Amico 77
Via Silvio D'Amico 77
Rome, RM 00145
Italy

Dario Brandolini

University of Turin ( email )

Via Po 53
Torino, Turin - Piedmont 10100
Italy

Paper statistics

Downloads
139
Rank
172,018
Abstract Views
885