Fuzzy clustering of financial time series based on volatility spillovers

Abstract In this paper we propose a framework for fuzzy clustering of time series based on directional volatility spillovers. the case financial series, detecting clusters spillovers provides insights into market structure, which can be useful to both portfolio managers and policy makers. We measure directional—i.e. “From” “To” others—volatility with methodology generalized forecast-error variance decomposition. Then, weighted model grouping stocks similar degree By using approach, allow algorithm decide dimension spillover is more relevant clustering. Moreover, robust also proposed alleviate effect possible outlier stocks. apply analysis effects in Italian stock market.