Abstract:
While it is widely agreed that Purchasing Power Parity (PPP) holds as a long-run concept the specific dynamic driving the process is largely build upon a priori economic belief rather than a thorough statistical modeling procedure. The two prevailing time series models, i.e. the exponential smooth transition autoregressive (ESTAR) model and the Markov switching autoregressive (MSAR) model, are both able to support the PPP as a long-run concept. However, the dynamic behavior of real exchange rates implied by these two models is very different and leads to different economic interpretations. In this paper we approach this problem by offering a bootstrap based testing procedure to discriminate between these two rival models. We further study the small sample performance of the test. In an application we analyze several major real exchange rates to shed light on the question which model best describes these processes. This allows us to draw a conclusion about the driving forces of real exchange rates.
