Dynamic Common Properties of National Herd Behavior of Stock Markets

Abstract

This chapter examines herd behavior across national borders. A dynamic latent factor model with Gibbs sampling is used to decompose the national herd behavior into the world, regional, and country-specific components. Testing the daily data from 2000 through 2014 for 47 countries, we find that the impact of world factor on national herd behavior is short-lived. This study indicates that world and regional factors play a significant role in explaining the variations of national herd behavior, constituting 33% of the herding variability. The significance of world and regional components is likely to produce a biased herding estimator.