Jakob Söhl is a researcher in statistics. He is assistant professor at Delft University of Technology.Contact
His research areas are statistics for stochastic processes, Bayesian statistics, nonparametric statistics and shape restricted inference.
He works on statistical applications in finance, forensics and sports.
Jakob Söhl did a postdoc with Richard Nickl at the University of Cambridge.
He wrote his PhD thesis under the supervision of Markus Reiß at the Humboldt-Universität zu Berlin.
Address:Preprints and Publications
Delft Institute of Applied Mathematics
Delft University of Technology
2628 CD Delft, Netherlands
- Improving state estimation through projection post-processing for activity recognition in football. ArXiv:2102.03310, 2021. (M. Ciszewski, J. Söhl, G. Jongbloed)
- Noise fit, estimation error and a Sharpe information criterion. Quantitative Finance, 20(6):1027-1043, 2020. (with D. Paulsen)
- Bernstein - von Mises theorems for statistical inverse problems II: Compound Poisson processes. Electronic Journal of Statistics, 13(2):3513-3571, 2019. (with R. Nickl)
- Shear viscosity computed from the finite-size effects of self-diffusivity in equilibrium molecular dynamics. Journal of Chemical Theory and Computation, 14(11):5959-5968, 2018. (S.H. Jamali, R. Hartkamp, C. Bardas, J. Söhl, T.J.H. Vlugt and O.A. Moultos)
- Nonparametric Bayesian posterior contraction rates for discretely observed scalar diffusions. Annals of Statistics, 45(4):1664-1693, 2017. (with R. Nickl)
- Adaptive confidence bands for Markov chains and diffusions: Estimating the invariant measure and the drift. ESAIM P&S, 20:432-462, 2016. (with M. Trabs)
- High-frequency Donsker theorems for Lévy measures. Probability Theory and Related Fields, 164(1):61-108, 2016. (with R. Nickl, M. Reiß and M. Trabs)
- Uniform central limit theorems for the Grenander estimator. Electronic Journal of Statistics, 9(1):1404-1423, 2015.
- Option calibration of exponential Lévy models: Confidence intervals and empirical results. Journal of Computational Finance, 18(2):91-119, 2014. (with M. Trabs)
- Confidence sets in nonparametric calibration of exponential Lévy models. Finance and Stochastics, 18(3):617-649, 2014.
- A uniform central limit theorem and efficiency for deconvolution estimators. Electronic Journal of Statistics, 6:2486-2518, 2012. (with M. Trabs)
- Polar sets for anisotropic Gaussian random fields. Statistics and Probability Letters, 80(9-10):840-847, 2010.
Most of my papers are also available on arXiv.
- An ultrasonic sensor for human presence detection to assist rescue work in large buildings. ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, IV-4/W7, 135-140, 2018. (T. van Groeningen, H. Driessen, J. Söhl and R. Vôute)
- Central limit theorems and confidence sets in the calibration of Lévy models and in deconvolution. Humboldt-Universität zu Berlin, 2013.
- Jakob Söhl teaches Probability and Statistics in the Bachelor Computer Science at Delft University of Technology.
- He teaches Statistics for Stochastic Processes as a Dutch national Master course. In the Winter Semester 2020/21 he lectured Statistics for Stochastic Processes as a Master course at the Berlin Mathematical School. Previously he gave this lecture as a Master course at the University of Cambridge.
- In the Winter Semester 2020/21 he taught the Bachelor course Statistical Methods for Data Science at the Berlin Mathematical School.
- From 2016 to 2020 Jakob Söhl taught Advanced Statistics to mathematics students at Delft University of Technology.
- In 2018/19 and 2019/20 he gave the lecture Probability and Statistics as part of the Bachelor Applied Earth Sciences in Delft.
- Jakob Söhl has been co-organising the Probability and Statistics Seminar in Delft from 2017 to 2020.
- He has been co-organising the Dynstoch 2019 workshop in Delft on 12-14 June 2019.
Links: Jakob Söhl, People in Statistics, Statistics Group