Publications

Peer-reviewed articles

  • A. Archimbaud, Z. Drmač, Klaus Nordhausen, U. Radojičić and Anne Ruiz-Gazen. Numerical Considerations and a New Implementation for Invariant Coordinate Selection. SIAM Journal on Mathematics of Data Science 5(1), 97-121. 2022
  • U. Radojicic, K. Nordhausen and J. Virta. Large-sample properties of blind estimation of the linear discriminant using projection pursuit. Electronic Journal of Statistics 15(2), 6677-6739. 2021
  • U. Radojičić, N. Lietzen, K. Nordhausen and J. Virta. Dimension estimation in two-dimensional PCA. In Proceedings of the 12th International Symposium on Image and Signal Processing and Analysis, pages 16-22. 2021.
  • K. Nordhausen, U. Radojičić. Least absolute value. To appear in Encyclopedia of Mathematical Geosciences. Encyclopedia of Earth Sciences Series. Springer, Cham.
  • U. Radojičić, K. Nordhausen and H. Oja. Notion of information and independent component analysis. Applications of Mathematics 65(3), 311-330. 2019.
  • U. Radojičićand K. Nordhausen. Non-Gaussian component analysis: testing the dimension of the non-Gaussian subspace.  In Maciak, M., Pestas, M. and Schindler, M. (editors) “Analytical Methods in Statistics. AMISTAT 2019”, 101-123, Springer, Cham. 2019
  • R. Scitovski, U. Radojičić and K. Sabo. A fast and efficient method for solving the multiple line detection problem. Rad HAZU, Matematičke znanosti. 23, 123 – 140. 2019
  • U. Radojičić, R. Scitovski and K. Sabo. A Fast and Efficient Method for Solving the Multiple Closed Curve Detection Problem, ICPRAM(2019), 269-276. 2019

 

Submitted manuscripts

  • J. Möttönen, K. Nordhausen, H. Oja and U. Radojičić. The Asymptotic Properties of the One-Sample Spatial Rank Methods, 2022
  • U. Radojičić, N. Lietzen, K. Nordhausen and J. Virta. Order Determination for Tensor-valued Observations Using Data Augmentation. arXiv preprint arXiv:2207.10423, 2021
  • U. Radojičić, K. Nordhausen and J. Virta. Kurtosis-based projection pursuit for matrix-valued data. arXiv preprint arXiv:2109.04167, 2021.

 

Theses

  • U. Radojičić, “Non-Gaussian Feature Extraction for Complex Data“, Doctor Thesis, Vienna University of Technology, 2021.
  • U. Radojičić, “Stochastic Optimization and Adaptive Annealing Method“, Master Thesis, J. J. Strossmayer University of Osijek, 2017.
  • U. Radojičić, “Calculus of Variations and Its Applications“, Bachelor’s Thesis, J. J. Strossmayer University of Osijek, 2015.

 

Software

  • J. Virta, C. Koesner, B. Li, K. Nordhausen, H. Oja and U. Radojičić. tensorBSS: Blind source separation methods for tensor-valued observations. R package version 0.3.8. CRAN.R-project.org/package=tensorBSS. 2021.
  • K. Riemer, G. Frahm, K. Nordhausen and U. Radojičić. shapeNA: M-estimation of shape for data with missing values. R package version 0.0.2. CRAN.R-project.org/package=shapeNA. 2021.