Chapter I - A HISTORY OF THE IDEA OF UNOBSERVED COMPONENTS IN THE ANALYSIS OF ECONOMIC TIME SERIES

Abstract

The idea that a time series may best be viewed as being composed of several unobserved components plays a fundamental role in certain areas of application. In the statistical literature, dealing with the analysis of economic time series, it is common practice to classify the types of movements that characterize a time series as trend, cyclical, seasonal, and irregular. This chapter reviews the literature on the trend, cyclical, seasonal, and irregular model to gain some understanding of the current uses of the model. It discusses the origins of the model and how it came into use in economics. The notion of unobserved components appears to have become common in economics during the period 1825–1875. In tracing the history of this idea, it is important to keep in mind the nature of scientific inquiry during this period, in particular the role of statistics. Moreover, the adoption of the model of unobserved components and separate periodicities in meteorology was important for the development of the model in economics.