Servicenavigation

Statistik: mehr als Erbsen zählen

Sie sind hier:

Dr. Burim Ramosaj

Mathematische Statistik und industrielle Anwendungen

Kontakt

Logistikcampus,
Raum A3.07
0231 755 - 90376
Fakultät Statistik
Technische Universität Dortmund
44221 Dortmund


Teaching

Summer Term 2020

 

Summer Term 2019

Brief CV

  • Since 08/2020 : Post-Doctoral Researcher at the Faculty of Statistics, Technical University of Dortmund. Research on Significant Variable Selection with Random Forest. Funded by the Ministry of Culture and Science of the state of NRW (MKW NRW) through the funding programme KI-Starter.
  • 07/2020: Graduation Dr. rer. nat., Faculty of Statistics, Institute of Mathematical Statistics and Applications in Industry, Technical University of Dortmund. Title of Dissertation: Analyzing Consistency and Statistical Inference in Random Forest Models.
  • 04/2019-07/2020: Research Assistant and Doctoral Student at the Faculty of Statistics, Technical University of Dortmund.
  • 04/2019-07/2020: Research Assistant and Doctoral Student at the Faculty of Statistics, Technical University of Dortmund.
  •  03/2017  - 03/2019: Research Assistant and Doctoral Student at the Institute of Statistics, University of Ulm.
  •  08/2015 - 08/2016: Studies of Mathematics (M.Sc.) at Syracuse University, NY, USA. Double degree program. 
  •  03/2014 - 02/2017: Studies of Mathematics and Management (M.Sc.) at the University of Ulm. Title of Master thesis: Multiple Imputation for the Daimler Financial Services AG.
  •  10/2010 - 01/2014: Studies of Mathematics and Management (B.Sc.) at the University of Ulm. Title of Bachelor thesis: First Passage Times of the Brownian Motion.

Research Interests

  • Asymptotic and Non-Parametric Statistics
  • Non-Parametric Classification and Regression
  • Statistical Inference with Machine Learning Methods
  • Missing Value Imputation

Publications and Preprints

  • Ramosaj, B., Interpretable Machines: Constructing Valid Prediction Intervals with Random Forest. (2021), preprint arXiv: 2103.05766
  • Ramosaj B. and Pauly M., Consistent and Unbiased Variable Selection using Random Forest Permutation Importance. (2021), preprint arXiv: 1912.03306
  • Yayla M., Günzel M., Ramosaj B. and Chen J.J. Universal Approximation Theorem of Fully Connected Binarized Neural Networks. (2021). preprint arXiv: 2102.02632
  • Ramosaj B. Analyzing Consistency and Statistical Inference in Random Forest Models. Dissertation (2020). http://dx.doi.org/10.17877/DE290R-21444
  • Amro L., Ramosaj B., Pauly M. Asymptotic based bootstrap approach for matched pairs with missingness in a single-arm. (2020). Biometrical Journal (to appear)

Memberships

  • Member of the American Mathematical Society
  • Mentor of the Friedrich-Ebert Foundation (Former head of scholarship holders of the Friedrich-Ebert Foundation, Group UIm)