The typical spectral shape of an economic variable: a visual guide

Abstract

Granger (1966) describes how the spectral shape of an economic variable concentrates spectral mass at low frequencies, declining smoothly as frequency increases. His empirical exercise focused on the evidence obtained from a handful of series. In this article, I focus on a broad range of economic variables to investigate their spectral shape. After performing several exercises, the results show that Granger’s assertion holds more often than not. Especially fuzzy cases are those that exhibit some degree of transition to a different regime, as are those estimated with a very short bandwidth.

Keywords:

• frequency domain
• spectral analysis
• nonparametric econometrics

JEL Classification:

• A20
• C02
• C14
• C18
• E32

Acknowledgements

I thank the comments and suggestions of Yan Carrière-Swallow, Consuelo Edwards, Carlos P. Medel and Pablo Pincheira. This article is a revised version of Working Paper 719, Central Bank of Chile. The views expressed in this article do not necessarily represent those of the Central Bank of Chile or its authorities. This article is dedicated to the memory of Cristóbal Kaufmann.

Notes

¹ For a comprehensive use of the spectrum for economic time-series analysis, see Granger and Watson (1984) and Hamilton (1994).