Why use Markov-switching models in exchange rate prediction?

A large amount of empirical studies finds the superiority of the regime-switching model in generating the process of exchange rates and in forecasting future exchange rates. This paper justifies the use of Markov-switching models by showing that this kind of time series process is consistent with the most popular exchange rate regime in the world - the dirty floating exchange rate regime. The theoretical implication of exchange rate determination indicates that a higher probability of a central bank's future interventions raises the rational expectations discrepancy between the exchange rate and its fundamentals, even though the bank does not step in the foreign exchange market during that period.