Nowcasting using real-time data

Policymakers increasingly rely on real-time measures of economic activity to inform decisions, yet official statistics are typically available only with substantial delay.

The aim of this research is to develop a methodology to extract information from high-frequency alternative indicators to produce weekly estimates of official monthly statistics in real time.

The methodology is applied to extract the information of Revolut payments data to build weekly trackers of retail sales and monthly GDP, and also to build a tracker for vacancies using online job advertisements.

Building on real-time tracking and nowcasting models, our approach addresses two challenges inherent in alternative indicators: the absence of seasonal adjustment and the prevalence of outliers. We incorporate seasonal components directly into the model, allowing the seasonal structure of low-frequency data to inform high-frequency proxies, and employ fat-tailed distributions to mitigate the influence of large, infrequent shocks.

The unobserved components model developed for this project enables us to extract information from seasonally unadjusted high-frequency data that is useful for producing accurate real-time estimates of low-frequency economic statistics. The proposed framework accommodates several features of real-time alternative indicators: they are typically available over a shorter time span than official statistics, exhibit both intra-year and intra-month seasonality, and are noisier than the series they are intended to track.

Our empirical exercises highlight the value of debit-card spending data and online job advertisements for tracking monthly statistics such as retail sales, GDP, and vacancies. This research provides strong support for the usefulness of alternative data sources to provide earlier estimates of key UK monthly economic variables.

Galvão, A. ‘Nowcasting using real-time data’ ESCoE Webinar, 29 January 2026.

De Polis, A., Galvão, A., Petrella, I. ‘Tracking weekly activity using new data sources‘ ESCoE Discussion Paper Series, ESCoE DP 2025-19, 28 November 2025.