Herding Behaviour and Contagion Effect: Application of Multivariate Dynamic Conditional Correlation GARCH Model
Paper presented at the Fourth International Conference on Business Analytics and Intelligence, IISc Bangalore, 2016
15 Pages Posted: 27 Aug 2018 Last revised: 2 Sep 2018
Date Written: December 20, 2016
Abstract
Maiden attempt is made to compare unconditional and conditional correlations to find out the role of herding behaviour on contagion effect in emerging Asian stock markets which are part of G-20 nations. Daily log market returns for the period from 1st April 2004 to 30th January 2009 are considered and Multivariate Dynamic Conditional Correlations are used for different study periods. It is found that simple correlation highlights the relationship between markets but fails to capture the contagion effect. On the other hand, M-DCC GARCH identifies the existence of contagion effect due to herding behaviour during the second phase of crisis. It is suggested to have early warning system at national and international level to alert the investors and countries for the possible effect of herding at the time of crisis.
Keywords: Herding Behaviour, Contagion, US Subprime Crisis, Emerging Stock Markets, M-DCC GARCH Model
Suggested Citation: Suggested Citation