Stock market volatility prediction using possibilistic fuzzy modeling

This paper addresses stock market assets return volatility forecasting and possibilistic fuzzy modeling. A recursive possibilistic fuzzy modeling (rPFM) approach is suggested to deal with the identification of systems affected by outliers and noisy data due to the use of memberships and typicalities to cluster data. Since financial markets are affected by news, expectations and investors psychology, the development of robust methodologies such as rPFM is essential for market participants. The model is evaluated for volatility forecasting with jumps using intraday data of different stock markets. Results indicate that rPFM is a potential tool for volatility forecasting and outperforming some traditional recursive fuzzy and neural fuzzy models.