The Forecasting Performance of Regime-switching Models of Speculative Behaviour

The aim of this study is to test whether a regime-switching model implied by the theory of periodically collapsing speculative bubbles is able to capture the observed dynamics of the exchange rates of the major currencies against the US dollar. To achive our goal, we initially estimate alternative bubble measures for the exchange rates under scrutiny by means of typical models for the fundamental values of exchange rates. We assume that the bubble can be in either the survival state or the collapse state. Moreover, the probability equations that governs the classification of regimes is allowed to be a function of the bubbles. Our next step is to determine the optimal regime-switching model that captures the stochastic properties of our data which will betested against the benchmark random walk model in an out of sample mode by employing typical tests for equal predictive ability. This model gives rise to the development of suitable trading strategies that will facilitate international investors to achieve profit maximization. Finally, we evaluate the reliability of the forecasts our regime-switching model by means of Monte Carlo simulations.