The analysis of cycles in macroeconomic time series began in earnest in the 1870s with the sunspot and Venus theories of William Stanley Jevons and Henry Ludwell Moore and the rather more conventional credit cycle theory of Clément Jugler (see Morgan, 1990, Chapter 1). Secular, or trend, movements were first studied somewhat later, with the term ‘trend’ only being coined in 1901 by Reginald Hooker when analysing British import and export data (Hooker, 1901). The early attempts to take into account trend movements, typically by detrending using simple moving averages or graphical interpolation, are analysed by Klein (1997), while the next generation of weighted moving averages, often based on actuarial graduation formulae using local polynomials, are surveyed in Mills (2011, Chapter 10). The first half of the twentieth century saw much progress, both descriptive and theoretical, on the modelling of trends and cycles, as briefly recounted in Mills (2009a), but it took a further decade for techniques to be developed that would, in due course, lead to a revolution in the way trends and cycles were modelled and extracted.
You are not authenticated to view the full text of this chapter or article.
Get access to the full article by using one of the access options below.