This thesis contains four chapters that cast new light on the ability of professional analysts and statistical models to assess economic growth in the current quarter (nowcast) and its development in the near future. This is not a trivial issue. An accurate assessment of the current state of the economy is important as starting point for medium-term forecasts, especially during times of heightened volatility, such as the recent financial crisis.

Nowadays, practitioners have a wealth of statistical model to choose from; but which one should they use? Can statistical models be modified to improve their forecasting accuracy? What are the gains from combining the forecasts of different statistical models? Did the financial crisis change the forecasting performance of statistical models relative to professional analysts? Can practitioners use the near-term outlook of professional analysts to improve the forecasting accuracy of statistical models? This thesis gives answers to these questions, providing new insights of interest to both academics and practitioners. Central to this research is the construction of a new dataset, comprised of the near-term economic growth forecast of professional analysts, and the monthly indicators available when analysts made their forecasts.

, , , , , , ,
J. Swank (Job)
Erasmus University Rotterdam
Erasmus School of Economics

de Winter, J. (2016, December 23). Nowcasting GDP Growth: statistical models versus professional analysts. Retrieved from