Economic Forecasting

The goal of economic forecasting is to produce reliable estimates of how economic conditions will develop. These estimates feed into budget decisions by governments, monetary policy decisions by central banks, and planning decisions by businesses. A forecast is never certain, so communicating the uncertainty around predictions is as important as the predictions themselves.

The work involves selecting a model, gathering the relevant data, and updating it as new data arrives. Common tasks include tracking indicator data such as consumer prices, labour market statistics, and trade figures, building or maintaining econometric models, and writing up results in reports or briefings. Forecasts are often produced on a regular schedule, such as quarterly or annual outlooks.

The main tools are R, Python, Stata, and EViews. Time-series methods such as ARIMA, VAR models, and state-space models are widely used. Machine learning methods are increasingly used alongside traditional econometric approaches, though interpretability remains important given that results must often be explained to non-technical audiences.

Economic forecasting projects appear most often at central banks such as DNB, government bodies such as the CPB, research institutes, and banks. In the Netherlands, the CPB's macro-economic forecasts feed directly into the national budget process. The project connects to Economic Researcher, Policy Analyst, and Risk Manager roles.