CMStatistics 2021: Start Registration
View Submission - CFE
A0457
Title: Weekly economic index for Belgium: Design and validation Authors:  Arno De Block - Vrije Universiteit Brussel (Belgium) [presenting]
Feliciaan De Palmenaer - Universiteit Gent and Vrije Universiteit Brussel (Belgium)
Kris Boudt - Vrije Universiteit Brussel and VU Amsterdam (Belgium)
Abstract: GDP growth is the key indicator for measuring the present state of the economy. Policymakers, firms, and investors closely monitor it to optimize their economic decision-making. But while decisions must be made in real-time, official GDP figures are only observed quarterly and with a substantial publication lag. Exploiting timely available data, one can predict the present state of the economy using econometric models, often referred to as nowcasting. The goal is to create a nowcast for Belgium on a weekly basis that tracks the economic activity. This Belgian Weekly Economic Index combines three data sources: banking data, electricity load data and media sentiment data. The banking data consists of sectoral aggregates of in- and outgoing cashflows of small to medium companies. The industry mainly dominates the electricity load, while economic sentiment can be extracted from newspaper articles. Combining these heterogeneous data sources using econometric models will lead to a weekly tracker of the Belgian economy.