Identification of Cyclical Phases : A Dynamic Factor - Markov Switching Model for India

The synthesis of dynamic factor model of Stock and Watson (1989) and the regime switching model of Hamilton (1989) proposed by Diebold and Rudebusch (1996) potentially encompasses both features of business cycles identified by Burns and Mitchell (1946) - co-movement among economic variables through the cycle, and non-linearity in its evolution. This paper estimates the dynamic factor-markov switching model for Indian data and examines whether both these features of business cycles are empirically relevant in the Indian context. The evidence clearly indicates the existence of common factor whose dynamics are well characterized by a 2-state Markov switching model with the duration of low-growth state being higher than that of the high-growth state. The usefulness of this model in identifying the growth cycle turning points through a statistical prediction algorithm is also highlighted