Modeling Business Cycles In The Romanian Economy Using The Markov Switching Approach

I use the Markov Switching AR approach to model the business cycles in Romanian economy for the 1991-2008 period using monthly data on industrial production. The time series used allows for a comparison with previous dating of Romanian business cycles. Generally, the MS-AR performs well, confirming the previous finding about turning points in business cycles during the transition period. At the same time, it suggests that the ongoing recession started earlier than conventionally thought and that it may last more than a year and a half.