Beauty and Ugliness of Aggregation over Time: A Survey

Abstract

This paper delivers an up-to-date literature review dealing with aggregation over time of economic time series, e.g. the transformation of highfrequency data to low frequency data, with a focus on its benefits (the beauty) and its costs (the ugliness). While there are some benefits associated with aggregating data over time, the negative effects are numerous. Aggregation over time is shown to have implications for inferences, public policy and forecasting.

A general property for time aggregation

“…Put another way, when realisations are generated from a timeaggregating model then samples of aggregated realisations still have the same model structure. The substantial literature on this topic is thoroughly surveyed by Silvestrini and Veredas [2008] and more recently by Mamingi [2017]. Almost all papers examine linear models within the general vector autoregressive integrated moving average class, see Tsai and Chan [2005], Foroni et al [2018] and many others.…”