The introductory course to robust multivariate statistics will take place on Monday 22 May. Please register here if you want to attend.

Robust estimation of multivariate location and scatter 

In this course, we will discuss two popular methods for robust estimation of multivariate location and scatter: S-estimators and the minimum covariance determinant estimator.

In part 1 of the course, we will introduce these methods, explain their relation with M-estimators, and discuss their robustness and asymptotic properties.

In part 2 of the course, we will discuss the mathematical techniques that can be used to rigorously derive the properties and features of these methods that have been discussed in part 1.

In part 3 of the course, we illustrate their implementation in R and how to use them in practice.

Speakers

Dr. Rik Lopuhaä finished his Master’s degree in Mathematical Statistics and Probability Theory at the University of Amsterdam in 1986, with an MSc thesis on shape constrained nonparametric statistics under the supervision of Piet Groeneboom. He received his PhD from Delft University of Technology in 1990, with a PhD thesis on multivariate statistical methods with high breakdown point under the supervision of Peter Rousseeuw. After his PhD he joined the Statistics group within the Delft Institute of Applied Mathematics at Delft University of Technology.

His main research interest is asymptotic theory of statistical methods, in particular for shape constrained nonparametric methods and for multivariate robust methods. He has several contributions on robust methods for multivariate location and scatter, ranging from the computation of breakdown points to robust and asymptotic properties of S-estimators, tau-estimators, MM-estimators, re-weighted estimators, and the MCD.

Dr. Valentin Todorov served for 15 years as a Senior Management Information Officer with the United Nations Industrial Development Organization (UNIDO) and after retiring from UNIDO works as an independent consultant on projects with the Joint Research Center (JRC) of the European Commission, the World Intellectual Property Organization (WIPO) and the Gulf Organization for Industrial Consulting (GOIC). He received a doctoral degree in statistics from Vienna University of Technology, Austria. His research activities include multivariate data analysis, computational statistics, robust statistics, official statistics, and statistical information systems.

He is an elected member of the International Statistical Institute (ISI), a member of the International Association for Statistical Computing (IASC) since 1991, the Austrian Statistical Society (ÖSG) and the Italian Statistical Society (SIS). Valentin Todorov has authored one book, book chapters and many scientific articles in mathematical statistics and statistical information systems. He has developed and maintains at CRAN several packages for the statistical software environment R and actively promotes R, especially in official statistics, by organizing and participating in trainings and technical cooperation projects in developing and emerging industrialized countries.