Geographical Diversification with a World Volatility Index

This paper proposes a new ‘World Volatility Index’, coined WVIX, by constructing the first index that approximates the aggregate volatility level of the G20 countries. The empirical analysis makes use of the factor dynamic conditional correlation model – with an automated methodology to detect the number of factors – in order to (i) sum up the information contained in the implied volatility indexes belonging to the US, the UK, the Eurozone, Japan and emerging countries, and (ii) examine the time-varying correlation between them. The results reveal that the WVIX evolves around 22%, but its activity can vary sharply depending on its exposure to various sources of geographical risks (e.g. the latest 2010-11 European debt crisis). Thus constructed as an early warning device, the methodology behind the WVIX can be replicated by market practitioners to datasets that better suit their needs.