A Steady-State Approach to Trend/Cycle Decomposition

In this paper, we present a new approach to trend/cycle decomposition. The trend of an integrated time series is estimated using the conditional expectation of the steady-state level of the series. Given a nonlinear forecasting model, this steady-state approach can differ in important ways from the related long-horizon forecast decomposition proposed by Beveridge and Nelson (1981). We use generated data from nonlinear regime-switching processes to demonstrate the advantages of the steady-state approach. We then apply the steady-state approach to estimate the trend and cycle of U.S. real GDP implied by a regime-switching forecasting model. Our findings portray a very different picture of the business cycle than implied by standard linear forecasting models.