• Behavioral Finance
• Cross-sectional volatility
• Herding
• Kalman Filter

Herd behavior is widely believed to play a crucial role in financial markets and particularly when the market is in stress. This work analyses the phenomenon of herd behavior from both a theoretical and an empirical point of view. We apply the approach by Hwang and Salmon (2004), based on the cross-sectional standard deviations of the betas, to analyse herd behavior in the Italian Stock Exchange in the period January 1998 - December 2012. We find that herd behavior towards the market portfolio is significant and persistent, independently from and given the particular state of the market, and it shows a positive correlation with the FTSE MIB. Another remarkable result, given that herd behavior can lead to significant mispricing, is that herd behavior is never greater than the 40% of its maximum potential value during the sample period. Further, we examine herd behavior towards SMB and HML factors and find evidence of significant periods of herd behavior towards SMB and HML.