RESEARCH INTERESTS

• Statistical Boosting


• Generalised Geo-Additive Mixed Models

TEACHING

• Advanced Statistical Inference (Likelihood and Bayes) (WiSe2023/2024)


• Data Science: Statistik (SoSe2023, SoSe2024, SoSe2026)


• Multivariate Statistics (SoSe2026)


• Grundlagen Bayesianischer Statistik und Statistisches Lernen (WiSe2024/2025, WiSe2025/2026, WiSe2026/2027)


• Introduction to Statistical Programming (WiSe2026/2027)


• Praktikum Statistische Modellierung (SoSe2024, SoSe2025, SoSe2026)




For theses, a personal meeting is recommended, where also topics outside of the theses offered here are possible.

SOFTWARE

• mermboost - A novel gradient boosting framework for mixed models. [R pacakge via CRAN]

PUBLICATIONS

1. Knieper, L., Hothorn, T., Bergherr, E., & Griesbach, C. (2025). Gradient boosting for generalised additive mixed models. Statistics and Computing, 35(4), 84. DOI





2. Knieper, L., Kneib, T., & Bergherr, E. (2024). Spatial confounding in gradient boosting. In International Workshop on Statistical Modelling (pp. 88-94). Cham: Springer Nature Switzerland. DOI





3. Gluesing, L., Potts, S., Knieper, L., Giganti, F., Tsaur, I., Gassenmaier, S., ... & Uhlig, J. (2026). MRI-derived prostate volume assessment: manual segmentation versus two-axis ellipsoid approximation and its implications for PSA density. European Radiology, 1-12. DOI

CONFERENCE CONTRIBUTIONS

• IWSM 2026 Selective Function-on-Surface Regression for Cortical Neuroimaging via Gradient Boosting (Talk)


• IWSM 2025 Spatially aware gradient boosting towards sparser models (Talk)


• DAGStat 2025 Re-thinking spatial components in gradient boosting (Talk)


• CompStat 2024 Spatial Confounding in Gradient Boosting (Talk)


• IWSM 2024 Spatial Confounding in Gradient Boosting (Talk)


• CMStatistics 2023 A novel gradient boosting framework for generalised additive mixed models (Talk)


• IWSM 2023 A novel gradient boosting framework for generalised additive mixed models (Poster)


• Statistical Computing 2022 Gradient boosting for linear mixed effects logit models