We live in an era where data plays a central role in our decision-making. The amount of data collected by government institutes, commercial organizations and scientists has exploded in recent years. Filtering and interpreting relevant information from modern datasets is a challenge that occupies many data scientists, including Etienne Wijler, assistant professor of Econometrics and Data Science.
“Recently available datasets contain enormous potential for scientific discoveries and data-driven decision-making. Think, for example, of studying economic interactions based on global observations, combined in macroeconomic datasets, evaluating climate policy based on air quality measurements from hundreds of monitoring stations, or providing proactive medical care through frequent monitoring of health indicators of thousands of patients via smartphones/smartwatches,” explains Wijler.
Wijler indicates that the analysis of the datasets in the aforementioned examples is complicated by the combination of two unique properties. Wijler: “These datasets are difficult to handle because they often contain large amounts of time series data, whose characteristics also change over time. For instance, consider the heavily fluctuating trends we currently observe in commodity and energy prices, which also translate into significant price changes in derived products. There are many methods to extract reliable information from large collections of stable time series or small collections of unstable time series. However, these methods are not easily applicable when both characteristics occur simultaneously, while this combination is increasingly common in modern datasets.”
Wijler will develop specialized techniques to enable scientists to reliably extract information from such challenging datasets, and to quantify the uncertainties in the analysis. Given the many disciplines working with large datasets in structurally changing environments, such as economics, climate science and health sciences, Wijler expects many interesting applications to emerge from his research. With the help of the grant, Wijler will also work to make his new methods known outside the traditional boundaries of the econometrics domain. Additionally, Wijler will personally work on an application close to his heart. Wijler: “Using the methods I will develop, I want to analyze the impact of regional variations in greenhouse gas emissions on climate change, to provide an improved framework for developing and quantifying risks in climate policy.”
About the Veni Grant
The NWO Veni grant, of up to 320.000 euros, is awarded to excellent researchers who have recently obtained their PhD, to conduct independent research and develop their ideas for a period of three years. Laureates are at the start of their scientific career and display a striking talent for scientific research.
Two researchers of VU School of Business and Economics have secured a Veni grant during this round. Econometrician Marina Friedrich will use her grant to make econometric methods more suitable and accessible for climate researchers, to study climate problems.