We introduce a maximally flexible Bayesian method for detecting structural breaks in panel data with applications in climate policy. Read more
Recommended citation: Konrad, L.D., Vashold, L., Crespo Cuaresma, J. (2026). "Bayesian Indicator-Saturated Regression" arXiv.
This paper presents an efficient algorithm to detect maximally influencial in linear regression and proposes consistency criteria for this to hold in double/debiased ML. Read more
Recommended citation: Konrad, L.D., Kuschnig, N. (2026). " Finding Most Influential Sets." Proceedings of the International Conference on Machine Learning (ICML).
This paper establishes the distribution of maximally influencial sets in linear models, allowing to test for excessive influence. Read more
Recommended citation: Konrad, L.D., Kuschnig, N. (2026). "Testing Most Influential Sets." Proceedings of the International Conference on Learning Representations (ICLR).
This paper investigates the effects of subsidies in basic research by means of a general equilibrium model. Read more
Recommended citation: Konrad, L.D. (2024). "Basically Trapped - A General Equilibrium Approach to the Undersupply of Basic Research in Free Markets" arXiv.
This paper develops a novel Bayesian prior distribution that approximates exact moment conditions while maintaining computational tractability. Read more
This paper analyzes macroeconomic risk by incorporating conditional skewness into growth forecasting models. Read more
This paper attempts to estimate counterfactual deforestation rates for evaluating REDD+ forest conservation programs using machine learning methods. Read more
This paper develops a novel methodology for testing influential subsets in linear regression. Read more
This work introduces an R package for Bayesian indicator saturated models, enabling robust detection of outliers and structural breaks using non-local priors. Read more