King's College London

Research portal

A cluster driven log-volatility factor model: a deepening on the source of the volatility clustering

Research output: Contribution to journalArticle

Original languageEnglish
JournalQuantitative Finance
Early online date14 Nov 2018
DOIs
Publication statusE-pub ahead of print - 14 Nov 2018

King's Authors

Abstract

We introduce a new factor model for log volatilities that considers contributions, and performs dimen- sionality reduction, at a global level through the market, and at a local level through clusters and their interactions. We do not assume a-priori the number of clusters in the data, instead using the Directed Bubble Hierarchical Tree (DBHT) algorithm to fix the number of factors. We use the factor model to study how the log volatility contributes to volatility clustering, quantifying the strength of the volatility clustering using a new non parametric integrated proxy. Indeed finding a link between volatility and volatility clustering, we find that a global analysis reveals that only the market contributes to the volatil- ity clustering. A local analysis reveals that for some clusters, the cluster itself contributes statistically to the volatility clustering effect. This is significantly advantageous over other factor models, since it offers a way of selecting factors in a statistical way, whilst also keeping economically relevant factors. Finally, we show that the log volatility factor model explains a similar amount of memory to a Principal Components Analysis (PCA) factor model and an exploratory factor model.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454