Crowd predictions of key economic indicators with YouGov and the Danish Finance Union
Crowd Prediction•1 minute read
There is a continuous need for monitoring and forecasting household credit, debt and savings, and unemployment for banks and governments. As traditional forecasts are often not accurate, this study tested the possibility of using crowd predictions to enable a more accurate forecast of fluctuations in economic indicators.
The project approach
In collaboration with the global public opinion and data company YouGov and the Danish Finance Union for Employees in Danish banks, we harvested predictions of fluctuations in credit, debts and savings 1, 3, 6 and 12 months ahead from the public, financial experts and bank employees in the Copenhagen Region. The analysis used 'Mean Absolute Percentage Error' (MAPE) to measure forecast accuracy (Hyndman and Koehler, 2006).
- Data from 1,239 citizens, 78 frontline employees, and 17 experts.
- 3.18% in Mean Absolute Percentage Error (MAPE) of predictions from frontline and experts when compared with the actual outcome.
- Financial experts and bank employees in large Danish banks can predict fluctuations in debt, savings, credit, and unemployment accurately 1 – 6 months ahead in the Copenhagen area.
- Crowd predictions can encourage proactive decision-making for banks and governments.
- Mindpool's Head of Research, Carina Hallin (2019), Collective Intelligence Unit, Copenhagen Business School. www.researchgate.net/publication/335795056_Testing_Smart_Crowds_for_the_Economy