Prof. Dr. Markus Löcher
Department of Business and Economics
Professor of Mathematics und statistics
+49 30 30877-1443
Postal address
Hochschule für Wirtschaft und Recht Berlin
Badensche Straße 52
10825 Berlin
Visiting address
Campus Schöneberg
Building B,
Room B 2.45
Badensche Straße 50-51
10825 Berlin
PhD, physics, Ohio University, 1997
Research Scientist, Georgia Institute of Technology, 1997-1999
Scientist, Siemens Corporate Research, Princeton, NJ, 1999-2003
Founder and CEO, DataInsight, 2004-2007
Masters in statistics, Rutgers University, 2007
Lead Scientist, Sense Networks, New York, 2007-2011
Computational Statistics in R, machine learning, clustering, prediction
Paradoxe
Bayesian regularisation
Data Visualisation
Publishing of reports
Visualisation of data on maps
Multiarm bandits
Improving Tree Ensemble Methods
Multiscale data analysis
Explainable Machine Learning
Computational Statistics, R
Statistics on the map (spatial statistics)
Large data sets
Location data
Cell phone apps
Are SHAP values biased towards high-entropy features? Raphael Baudeu, Marvin Wright and Markus Loecher, ECML PKDD Workshop on XKDD, Grenoble (2022).
Approximation of SHAP values for Randomized Tree Ensembles, Loecher, Markus, Dingyi Lai and Qi Wu, CD-MAKE (2022).
Debiasing MDI Feature Importance and SHAP values in Tree Ensembles, Loecher, Markus, CD-MAKE (2022)
The Perils of Misspecified Priors and Optional Stopping in Multi-Armed Bandits, Loecher, Markus, Frontiers in Artificial Intelligence 4 (2021).
Unbiased variable importance for random forests, Loecher, Markus, Communications in Statistics - Theory and Methods (2020): 1-13.
Real-world insights on public charging demand and infrastructure use from electric vehicles Hardinghaus, Michael, Markus Löcher, and John E. Anderson. Environmental Research Letters 15.10 (2020): 104030. (Won an internal DLR price )
Rapid Polygon Lookup, M. Loecher and M. Kumar, Springer Communications in Computer and Information Science (2015)
RgoogleMaps and loa: Unleashing R Graphics Power on Map Tiles, M. Loecher and K. Ropkins, Journal of Statistical Software, vol. 63, no. 4 (2015).
Noise Sustained Patterns: Fluctuations and Nonlinearities, M. Löcher, World Scientific Lecture Notes in Physics, World Scientific (2003).
CitySense: Spatiotemporal Anomaly Detection, Joint Statistical Meeting (JSM 2009), Washington DC.
Noise enhanced propagation, J. F. Lindner, S. Chandramouli, A.R. Bulsara, M. Löcher, and W. L. Ditto, Phys. Rev. Lett. 81, 5048 (1998).
Control of high-dimensional chaos in systems with symmetry, M. Löcher and E. R. Hunt, Phys. Rev. Lett. 79, 63 (1997).
Spatiotemporal Stochastic Resonance in a System of Coupled Diode Resonators, M. Löcher, G. A. Johnson and E. R. Hunt, Phys. Rev. Lett. 77, 4698 (1996).
Columbia University, NY, USA
Princeton University, NJ, USA
Fachhochschule Münster
Ohio University, OH