Peter Rousseeuw
Peter J. Rousseeuw is a Belgian statistician known for his work on robust statistics and cluster analysis. He obtained his PhD in 1981 at the Vrije Universiteit Brussel, following research carried out at the ETH in Zurich, which led to a book on influence functions. Later he was professor at the Delft University of Technology, The Netherlands, at the University of Fribourg, Switzerland, and at the University of Antwerp, Belgium. Next he was a senior researcher at Renaissance Technologies. He then returned to Belgium as professor at KU Leuven, until becoming emeritus in 2022. His former PhD students include Annick Leroy, Hendrik Lopuhaä, Geert Molenberghs, Christophe Croux, Mia Hubert, Stefan Van Aelst, Tim Verdonck and Jakob Raymaekers.
Research
Rousseeuw has constructed and published many useful techniques. He proposed the Least Trimmed Squares method and S-estimators forrobust regression, which can resist outliers in the data.
He also introduced the Minimum Volume Ellipsoid and Minimum Covariance Determinant methods
for robust scatter matrices. This work led to his book Robust Regression and Outlier Detection with Annick Leroy.
With Leonard Kaufman he coined the term medoid when proposing the k-medoids method for cluster analysis, also known as Partitioning Around Medoids.
His silhouette display shows the result of a cluster analysis, and the corresponding silhouette coefficient is often used to select the number of clusters. The work on cluster analysis led to a book titled Finding Groups in Data.
Rousseeuw was the original developer of the R package cluster along with Mia Hubert and Anja Struyf.
The Rousseeuw–Croux scale estimator is an efficient alternative to
the median absolute deviation.
With Ida Ruts and John Tukey he introduced the bagplot, a
bivariate generalization of the boxplot.
His more recent work has focused on concepts and algorithms for statistical depth functions in the settings of multivariate, regression and functional data, and on robust principal component analysis. His current research is on visualization of classification and cellwise outliers.
For more information see a 2024 interview.