Document details

Time and cross-sectional differences in the tail behavior of Euro interest rate future returns

Author(s): Neumann, Christian

Date: 2017

Persistent ID: http://hdl.handle.net/10362/26209

Origin: Repositório Institucional da UNL

Subject(s): Extreme value theory; Tail index estimation; Optimal extreme sample fraction; Structural change test; Government bond futures; Market risk management; Monetary policy; Domínio/Área Científica::Ciências Sociais::Economia e Gestão; Domínio/Área Científica::Ciências Sociais::Economia e Gestão; Domínio/Área Científica::Ciências Sociais::Economia e Gestão


Description

As response to the financial crisis in 2007/08 and the European sovereign debt crisis, the ECB started to conduct expansionary monetary policy on an unprecedented scale. In this paper I investigate the development of tail risks in the euro interest rate market since the implementation of this unconventional monetary policy. The focus of the study is on futures on German government bonds, namely the Bund, Bobl and Schatz, which are among the most relevant securities in this market. The analysis covers three aspects. First, I investigate if the daily returns of the futures exhibit fat tails over the period from 1999 to 2016 and if there are differences among these securities with respect to tail risk, as measured by the tail index. Second, I analyze if the tail risks are non-constant over the considered time period. Third, I study if the tail index contains information beyond the conventional risk measure volatility and its implications for value-at-risk considerations. Anticipating the results, this paper presents significant evidence for fat tails in the return distribution of the Bund, Bobl and Schatz future. In contrast to expectations, the results indicate the highest tail risk for the short-term Schatz future and the lowest for the long-term Bund future. Differences in market liquidity might be a reason for this. Furthermore, I find comprehensive evidence for an increase in right tail risk for all three futures around 2008. This increase is most significant for the long-term Bund future. Surprisingly, evidence for a decrease in left tail risk is found, although with lower significance. Additionally, the analysis reveals that tail index contains information, which is not captured by volatility. Thus, the results suggest that the accuracy of value-at-risk estimates for different long and short positions can be improved by taking into account the tail index explicitly in the estimation process.

Document Type Master thesis
Language English
Advisor(s) Boons, Martijn
Contributor(s) Neumann, Christian
facebook logo  linkedin logo  twitter logo 
mendeley logo

Related documents

No related documents