The Typical Spectral Shape of an Economic Variable: A Visual Guide with 100 Examples

Granger (1966) describes how the spectral shape of an economic variable concentrates spectral mass at low frequencies, declining smoothly as frequency increases. Despite a discussion about how to assess robustness of his results, the empirical exercise focused on the evidence obtained from a handful of series. In this paper, I focus on a broad range of economic variables to investigate their spectral shape. Hence, through different examples taken from both actual and simulated series, I provide an intuition of the typical spectral shape of a wide range of economic variables and the impact of their typical treatments. After performing 100 different exercises, the results show that Granger's assertion holds more often than not. I also confirm that the basic shape holds for a number of transformations, time aggregations, series' anomalies, variables of the real economy, and also, but to a lesser extent, financial variables. Especially fuzzy cases are those that exhibit some degree of transition to a different regime, as are those estimated with a very short bandwidth.