Rate My Professor Klaus Nordhausen

Nordhausen's research centers on robust and nonparametric multivariate statistical methods, including blind source separation, independent component analysis, invariant coordinate selection, and techniques for high-dimensional, spatial, spatio-temporal, and tensor-valued data. These methods enable the extraction of meaningful signals from noisy datasets, with applications in industrial quality control, environmental monitoring, and medical research. Notable publications include 'Independent vector analysis – an introduction for statisticians' (2026, with M. Arvila, M. Sipilä, S. Taskinen), 'Order Determination for Tensor-Valued Observations Using Data Augmentation' (2026, with U. Radojičić, N. Lietzén, J. Virta), 'Kurtosis-based projection pursuit for matrix-valued data' (2025, with U. Radojičić, J. Virta), 'JADE for Tensor-Valued Observations' (2018, with J. Virta, B. Li, H. Oja), and 'Blind Source Separation Based on Joint Diagonalization in R: The Packages JADE and BSSasymp' (2017, with J. Miettinen, S. Taskinen). He has edited volumes such as 'Robust and Multivariate Statistical Methods: Festschrift in Honor of David E. Tyler' (2023, with M. Yi) and 'Modern Nonparametric, Robust and Multivariate Methods' (2015, with S. Taskinen). Nordhausen has supervised PhD students including Aurore Archimbaud, Markus Matilainen, and Joni Virta, and secured funding from the Academy of Finland for the project 'Signal recovery in noisy spatial data' amounting to 583,258 euros.