Abstract:
Infra-monthly time series have increasingly appeared on the radar of official statistics in recent years, mostly as a consequence of a general digital transformation process and the outbreak of the COVID-19 pandemic in 2020. Many of those series are seasonal and thus in need for seasonal adjustment. However, traditional methods in official statistics often fail to model and seasonally adjust them appropriately mainly since data of such temporal granularity exhibit stylised facts that are not observable in monthly and quarterly data. Prime examples include irregular spacing, coexistence of multiple seasonal patterns with integer versus non-integer seasonal periodicities and potential interactions as well as small sample issues, such as missing values and a high sensitivity to outliers. We provide an overview of recent modelling and seasonal adjustment approaches that are capable of handling these distinctive features, or at least some of them. Hourly counts of TARGET2 customer transactions and daily realised electricity consumption in Germany are discussed for illustrative purposes.