Decision Support; Risk and Marketing Analytics; Forecasting; Deep Learning and AI
Stefan’s work focuses on the analysis, support, and automation of managerial decision-making using empirical-quantitative methods. The modeling of customer behavior in marketing and the prediction of risk factors and profitability in the credit business are distinctive pillars of Stefan’s research. However, he is equally interested in other data intensive applications in the broad scope of management and economics including, for example, demand planning, financial forecasting, and algorithmic text analysis. An overarching research objective is to contribute toward the marriage of scalable (deep) machine learning algorithms and theory-grounded econometric methods for explanation and causation. Recent work on uplift models, which predict individual-level treatment effects such as the differential increase in customer spending due to a marketing action illustrate this approach and will remain an important area for future research.