Economies and markets are subject to substantial political and economic uncertainty, such as the covid19 pandemic (2020), Brexit (2020), the European debt crisis (2010-2014), and the financial crisis (2008). Assessing such risks for systems of countries or portfolios of assets across different markets (e.g. for pension and insurance funds) is important from an academic, regulatory, and professional perspective. Importantly, large risks may affect different parts of the economy, or different countries in a region, very differently. Good empirical models to support decision making therefore need to accommodate such heterogeneity in risk response to be useful for regulators and practitioners alike. A major drawback of most contemporary models used to assess extreme economic risks, also known as tail risks, is that they allow for heterogeneity risk response only in a highly restricted way. This may lead to an under- (or over)-estimation of systemic risk assessment. Both biases may result in substantial societal costs. This proposal introduces a new econometric methodology that allows for much more tail risk heterogeneity than existing models. The new approach can be used to describe the heterogeneity in extreme risks more accurately. The approach builds on new flexible multivariate distributions that accommodate tail heterogeneity, while at the same time retaining computational tractability. The latter is particular important if we consider big data applications in high dimensions, such as many industries across many countries, or many individual firms or assets. Extensions of the methodology also allow for heterogeneity in risk spill-over from one country/industry/firm to the next. The results will be useful both for academics, regulators, central banks, and practitioners and will be made publicly available via computer code and relevant applications.
|Effective start/end date||1/01/22 → 31/12/26|
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.