Recognizing business cycle turning points by means of a neural network

The latest, 1990–91 recession marks the ninth downturn in the U.S. economy during the past fifty years. There is scope for adding extensions to the methodology of monitoring such major economic fluctuations. The use of artificial neural networks is proposed here. For demonstration a case study is included. In it four key economic indicators are examined; viz., sales, production, employment and personal income. The growth rate movement common to these variables is represented by a state space model of dynamic systems theory. Their monthly time series data over 1965–1989 are simultaneously analyzed. The dates of business cycle peaks and troughs identified in the analysis agree closely with the official chronology.