A further note on the three phases of the US business cycle
- 1 July 2000
- journal article
- research article
- Published by Taylor & Francis in Applied Economics
- Vol. 32 (9) , 1133-1143
- https://doi.org/10.1080/000368400404272
Abstract
Using a number of alternative approaches, Sichel (1994) demonstrated evidence supporting the notion that the US business cycle is best characterized as having three distinct phases, viz. contraction, followed by rapid expansion during the early stages of the recovery phase, followed by a period of more normal expansionary growth, with the cycle then repeating itself. This contrasts with the more usual expansion/contraction, two phase characterization but is more in keeping with the original notion of the business cycle as conceived by Burns and Mitchell (1946). Here an alternative approach is employed for shedding light on this issue. Following the original suggestion of Hamilton (1989, 1990, 1991), a simple nonlinear, three phase, regime switching Markov model is compared against its simpler two phase version to determine which version is statistically more consistent with the business cycle historical evidence. The evidence seems to clearly support the three phase characterization and that this characterization yields interesting information on business cycle dynamics which is necessarily missed by the two phase model formulation.Keywords
This publication has 9 references indexed in Scilit:
- Do Leading Indicators Really Predict Australian Business Cycle Turning Points?Economic Record, 1997
- Dating and predicting phase changes in the U.S. business cycleInternational Journal of Forecasting, 1996
- Specification testing in Markov-switching time-series modelsJournal of Econometrics, 1996
- A Markov Model of Switching-Regime ARCHJournal of Business & Economic Statistics, 1994
- Inventories and the Three Phases of the Business CycleJournal of Business & Economic Statistics, 1994
- Cyclical Patterns in the Variance of Economic ActivityJournal of Business & Economic Statistics, 1993
- A Quasi-Bayesian Approach to Estimating Parameters for Mixtures of Normal DistributionsJournal of Business & Economic Statistics, 1991
- A New Approach to the Economic Analysis of Nonstationary Time Series and the Business CycleEconometrica, 1989
- Scoring the Leading IndicatorsThe Journal of Business, 1989