FURTHER EVIDENCE ON THE MONTE CARLO CYCLE IN BUSINESS ACTIVITY

This paper tests the null hypothesis that "cycles" in business activity are merely Monte Carlo cycles. It compares the actual distribution of expansionary and contractionary phase duration times with those which would be expected under the null hypothesis. If the probability of a phase turning point is constant during a small interval of time, rather than increasing, the duration times of upswings and downswings should be exponentially distributed. Goodness of fit statistics were calculated for each phase for the post World War II time series of: five coincident indicators, inverted unemployment rate, and production index. The evidence found here supports the Monte Carlo thesis.