‘Classical’ Techniques of Modelling Trends and Cycles

Abstract

Researchers studying the growth and cyclical behaviour of industrialised economies are immediately faced with the problem of separating cyclical fluctuations from longer-term, or secular, movements. The difficulties in doing this have been well appreciated for many years, but the traditional methods of trend and cycle decomposition were essentially ad hoc, being designed primarily for ease of computation without real regard for the statistical properties of the time series (or set of series) under analysis.