Predicting US business-cycle regimes

This article examines the use of logistic classification methods for the id entification and prediction of postwar U.S. business-cycle expansion and co ntraction regimes as defined by the National Bureau of Economic Research (N BER) reference turning-point dates. We present a coherent theoretical frame work for this task using measures of discriminatory information. The analys is encompasses model selection, parameter estimation, and classification de cision rules. We examine the performance of logistic procedures in reproduc ing the NBER regime classifications and in predicting one and three months ahead using leading-indicator variables. Our models are shown to provide su bstantially more accurate business-cycle regime predictions than Markov swi tching specifications.