Vol. 54 No. 3 (2025): Special Issue. In memorial: Fritz Leisch

This special issue of the Austrian Journal of Statistics is devoted to the memory of Friedrich "Fritz" Leisch who passed away after a serious illness a year ago, in April 2024. The idea for the issue was developed by a group of collaborators and friends of Fritz, consisting of Bettina Grün, Kurt Hornik, Torsten Hothorn, Theresa Scharl, and Achim Zeileis. Our aim was to compile contributions which honor Fritz' diverse scientific contributions to statistical computing, literate programming, cluster analysis and mixture models, statistical graphics, and applied statistics. Contributions were by invitation only and issued to a number of Fritz' co-authors. Bettina Grün and Theresa Scharl processed the special issue as Guest Editors. We would also like to thank the Editor of the Austria Journal of Statistics, Matthias Templ, and the Copy Editor, Klara Hruzova, for their support.

The special issue covers two contributions honoring Fritz' impact on reproducible research and literate programming by Roger Peng (University of Texas at Austin) and by Robert Gentleman (Dana Farber Cancer Institute), Antony Rossini (UCB and University of Washington), and Vincent Carey (Harvard Medical School), respectively. A contribution by Fritz Leisch and Torsten Hothorn (both at LMU Munich when drafting this in 2011) on inference for mixture models is finally published. In addition, Torsten Hothorn (University of Zurich) reflects on the reproducibility of the ten-year-old simulation study included in this work. Three contributions extend clustering methodology developed by Fritz and are accompanied by new R packages, available from the Comprehensive R Archive Network (CRAN). Dominik Ernst, Lena Ortega Menjivar, Theresa Scharl (all BOKU University), and Bettina Grün (WU Wien) discuss distance-based as well as model-based clustering methods for ordinal data; Matthias Medl, Ursula Laa (both BOKU University), and Dianne Cook (Monash University Melbourne) provide interactive exploration and visualization methods for market segmentation; and Lucas Sablica, Kurt Hornik, and Bettina Grün (all WU Wien) contribute to spherical and circular clustering in text mining. Fritz' general interest in different areas of statistics, in particular when useful for applied work, including robust, educational, and environmental statistics is reflected by the remaining contributions. Bernhard Spangl (BOKU University) investigates the robustification of the Kalman filter in a multivariate setting, Achim Zeileis (University of Innsbruck) presents different approaches for assessing measurement invariance and for detecting differential item functioning in the Rasch model along with their software implementation in R, and Gregor Laaha, Johannes Laimighofer, Nur Banu Özcelik (all BOKU University), and Svenja Fischer (Wageningen University) provide four case studies where accounting for heterogeneity based on domain knowledge improves the statistical modeling approach.

Bettina Grün and Theresa Scharl (Guest Editors)