Detta är ett uppsatsförslag hämtat från Nationella Exjobb-poolen. Klicka här för att komma tillbaka till samtliga exjobbsförslag.
Correlation measures in coordinated sell-off events
Recently we have seen market events when the correlations between apparently uncorrelated assets suddenly spikes. These 'nowhere-to-hide'-events are associated with substantial market corrections where a broad range of assets fall simultaneously. For a portfolio manager it is essential to be aware of the risk of such a correction and even try to estimate the probability of correlation spikes.
The purpose of this master thesis project is to make an extensive overview of possible methods to construct a risk measure for coordinated market sell-offs based on correlation forecasting. The starting point will be the Dynamic Equicorrelation Models for Large Correlations Matrices (DECO) by Robert Engel, based on an ARCH-approach. The most promising methods will be implemented and compared using daily data provided by RPM. Necessary requirement are a firm base in mathematical finance and programming, with strong skills in SQL and either Matlab, C#/.NET or any of the statistical analysis tools held in-house.
Informationen om uppsatsförslag är hämtad från Nationella Exjobb-poolen.