Identification And Prediction Of Currency Crisis: Markov Switching-Based Approach

This paper investigates ways of identifying and predicting currency crises in world-wide markets, with special focus on 1997 and 2008 currency crises. A novel Markov switching method is proposed for identifying currency crisis based on two states model, the turmoil state and tranquil state, which is the most suitable model considering the balance between model performance and computational demand. Compared with previous Markov switching currency crisis studies, the contribution of this paper comes from several ways. First, the dependent variable is different. While other papers use the exchange rate directly or the estimation of devaluation probability, this study uses the market pressure index calculated from nominal exchange rate and foreign reserves. Secondly, we allow different volatilities in different states, whereas other papers assume the same volatility in two states. Thirdly, our transition probabilities are constant rather than time-varying. The model shows evidence of state switching before crisis in many different currency markets. Lastly, we compare the Markov switching method with the widely used probit model which proposed an early warning system in terms of forecasting performance, and the empirical results show that the novel Markov switching method performs better than the probit model.