Multivariate regime switching GARCH-family models: inference, stability, filtering and forecasting

The collaboration seeks to 1) solve and develop the statistical inference of certain classes of Markov Switching GARCH-type models, including their structure, stationarity conditions, estimation methods, and asymptotic theory. 2) investigate volatility forecasting and impulse responses for a class of multivariate Markov Switching GARCH models and explore the role of the structure linkage and interdependencies among various financial variables in the appearance of Markov switches. The group has one co-authored application concerning the multivariate regime switching BEKK models, submitted to a well-recognized journal. The article proposed two estimation algorithms by using extended Kalman filters, derived from suitable state space representations of the proposed model. Both have been unable to collaborate effectively since due to either the heavy teaching duties or the heavy administrative roles. The ICMS Research-in-Groups programme will provide an excellent foundation to reinvigorate the international collaboration of Dr Cheng and Dr Cavicchioli to develop innovation in the context of multivariate regime switching GARCH models with applications in financial markets.