Master's degree > Economics: Nowcasting Model for GDP

• You are following an economics Master's program with a clear quantitative component.

• You have knowledge of - and preferably some experience with - software development and programming in R.

• You work accurately and in a structured manner, both with data and with code and documentation.

• You are analytically strong and know how to fathom complex issues and interpret data critically.

• You think proactively about improving existing methods and are open to exploring new ideas and techniques.

• You take responsibility for your own work: you plan, organize, and monitor your progress during the internship.

Because all communication during the internship - both oral and written - takes place in Dutch, a good command of the Dutch language is necessary.

A pleasant learning environment in which you:

• Applying your programming and analysis skills in a practical environment;

• Increasing your professional knowledge by collaborating with experienced colleagues;

• Applying your knowledge of econometric and statistical models based on economic indicators in practice;

• Gaining valuable work experience for a future career in data analysis or research.

In return for your commitment, you will receive an internship allowance of € 800,- gross per month (based on 40 hours per week).

Together with your internship supervisor, you will coordinate how you can work hybridly, with flexible working hours and the option to work partly from home on your internship assignment.

Do you get energy from building smart data models that predict economic growth even before all the figures are in? Do you love diving into data, discovering new data sources, and do you want to make a real impact on official GDP estimates through your analyses?
Then this internship in nowcasting and econometrics is exactly what you are looking for.

When estimating quarterly GDP (economic growth) figures, we currently use a relatively simple method for a number of industries: the development in the number of jobs serves as a proxy for economic activity. This is because direct and timely source information is lacking. A major consequence of this is that we often have to adjust the first estimates later when more complete data becomes available for the final estimate (approximately one and a half years later).

It is up to you to improve this methodology.

You will develop a more advanced model that allows us to more accurately estimate the economic activity of these industries shortly after the end of the reporting period. This is also known as "nowcasting." The goal is to be closer to the final realization and thereby reduce the number and magnitude of later adjustments.

An important part of your assignment is to investigate which data sources are available in a timely manner and can be used as input for this model. Subsequently, you will use historical data to simulate previous reporting periods. This allows you to analyze whether your proposed method actually performs better than the current approach, for example by looking at smaller adjustments.

Finally, you will develop a working production pipeline in R, so that the new method becomes practically applicable within the estimation process.