Publication

This paper proposes the Markov switching model to measure herding behavior in markets. The authors say that by means of time-varying transition probabilities, the model is able to link variations in herding behavior to proxies for sentiment or the macroeconomic environment. The authors use the model on US stock market data and find that during times of high volatility in the market, investors discriminate more strongly between single stocks than during tranquil times. Furthermore, this discrimination is greater than that implied by rational asset pricing models.