This paper puts forward a method to estimate average economic growth, and its associated confidence bounds, which does not require a formal decision on potential unit root properties. The method is based on the analysis of either difference-stationary or trend-stationary time series models, implementing the robust bootstrapping procedure advocated in Romano and Wolf (2001). Simulation evidence indicates the practical relevance of the method. It is illustrated on quarterly post-war US industrial production.

growth, robust testing, unit root
Estimation (jel C13), Simulation Methods; Monte Carlo Methods; Bootstrap Methods (jel C15), Time-Series Models; Dynamic Quantile Regressions (jel C22)
Tinbergen Institute Discussion Paper Series
Tinbergen Institute

Boswijk, H.P, & Franses, Ph.H.B.F. (2001). How Large is Average Economic Growth? Evidence from a Robust Method (No. TI 02-002/4). Tinbergen Institute Discussion Paper Series. Retrieved from