DATING TURNING-POINTS IN THE BUSINESS-CYCLE

I review and evaluate procedures for dating peaks and troughs of the business cycle. Five different methods are judged on their historical performance, ease of replication, clarity, timeliness, flexibility, and the validity of underlying assumptions. In addition, I examine the 1990-92 period in detail where two newer dating methods are especially insightful because they use statistically rigorous methods to derive turning point probabilities, and, thereby, the inherent uncertainty in the data can be recognized. I also conclude that additional research is needed on this topic to resolve some important controversies about business-cycle characteristics.