Abstract
This study revisits the detection of turning points for G-7 business cycles using a Markov switching model with latent information (MSLI). The MSLI model, championed by Otranto (2008), is useful in the situation in which the information is not available in real time, or is conditionally correlated with the states of the Markov chain, or when it is difficult to identify an observed information variable. The empirical results show that the time varying transition probability is helpful in signaling the business cycle conditions in the near future. The ability of the MSLI model to identify business cycle turning points is definitely better than that of the Markov switching (MS) model in the case of the US. For France, Italy, Germany and Japan, the MSLI model generates several false signals (viz, when the model predicts that there will be a recession, but one does not actually occur) and missed signals (i.e., when there is a recession, but the model fails to predict it) in dating turning points.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
