In this paper a dynamic bi-factor model with Markov switching is proposed to measure and predict turning points of the German business cycle. It estimates simultaneously the composite leading indicator (CLI) and composite coincident indicator (CCI) together with corresponding probabilities of being in recession. According to the bi-factor model, on average, CLI leads CCI by 3 months at both peaks and troughs. The model-derived recession probabilities of CCI and those of CLI with a lag of 2?3 months capture the turning points of the ECRI?s and OECD?s reference cycle much better than the dynamic single-factor model with Markov switching.
Forecasting turning points composite coincident indicator composite leading indicator dynamic bi-factor model Markov-switching