Herding, Trend Chasing and Market Volatility

Abstract

We introduce a heterogeneous agent asset pricing model in continuous-time to show that trend chasing, switching and herding all contribute to market volatility in price and return and volatility clustering, but their impact are different. On the one hand, the fluctuations of market price and return and the level of the significant autocorrelations (ACs) of the absolute and squared returns increase with herding and trend chasing based on long time horizon. On the other hand, the switching reduces the return volatility and an initial increase in switching reduces the price volatility and increases the level of the significant ACs, but the effect becomes opposite when the switching increases further. We also show that market noise plays more important role than fundamental noise on the power-law behavior.

Keywords: Heterogeneous beliefs, herding, switching, stability, volatility, stochastic delay differential equations

JEL Classification: C62, D53, D84, G12