Institute of Econometrics and Data Science

Subject

The Institute for Econometrics and Data Science is dedicated to the development and application of empirical methods. It is situated at the intersection of econometrics, machine learning, and empirical finance. The financial sector is very data-intensive, and an adequate understanding of these large datasets and their dependencies over time and cross-sectionally is essential. Econometric methods allow us to estimate the magnitude of economically relevant parameters and to make statements about estimation uncertainty. New methods are continually being developed to address the specific requirements and settings of current questions and datasets. Due to the richness of data in the financial sector, machine learning methods have been increasingly used in recent years to identify patterns in observations flexibly. An exciting field of research arises from the question of how this flexibility can be combined with established financial theories and used to support classical econometric methods.

Teaching

The Institute for Econometrics and Data Science emphasizes skills development in both theory and application. The goal is to enhance students' understanding of various econometric estimation methods and machine learning techniques. This is achieved through the integration of statistical software. Data Science is inherently interdisciplinary, researched, and utilized by various scientific disciplines. Skills in this field are among the most sought-after in current job postings. In competition with applicants from other scientific disciplines, students of economics and business administration can be particularly successful if they are able to combine methodological expertise with economic domain knowledge. Therefore, advanced courses also address the intersection of quantitative methods and empirical (financial) economic research. The objective of education at the Institute is to prepare students methodically for a challenging career in the private sector while also opening the possibility for an academic career.

Research

The research focuses of the Institute lie at the intersection of econometrics, machine learning, and empirical economic research. They include, among other things, simulation-based estimation methods and examining how machine learning techniques can be effectively used to support or test financial theory. Past and present research projects address, for example, extreme forms of sample selection within the framework of consumption-based models in asset pricing, estimate COVID-19 mortality, or analyze to what extent the approximation error in theory-driven return forecasts can be explained by machine learning methods. The data used in these analyses are diverse, ranging from macroeconomic time series at a quarterly frequency to trade data at the millisecond level. The Institute is involved in the DFG research group Financial Markets and Frictions.