Tuesday, March 4, 3.15 pm, 2025. Seminar in Statistics.

Covariance matrix estimation in a doubly multivariate model
Monika Mokrzycka, Institute of Plant Genetics, Polish Academy of Sciences
Abstract: The aim of the talk is to present methods of covariance matrix estimation by a symmetric positive definite separable structure. The estimation criterion is the entropy loss function. Such estimation is of interest for data where characteristics are measured repeatedly, e.g., in time or location. The approach can also be used to measure the discrepancy between two covariance matrix structures related to the study of the power of a test. However, modern real-world experiments produce high-dimensional data, which makes approximation via the entropy loss function challenging.
Location: Alan Turing.

Tuesday, April 8, 3.15 pm, 2025. Seminar in Mathematical Statistics.

Unsupervised linear discrimination using skewness
Joni Virta, Department of Mathematics and Statistics, University of Turku
Abstract: It is known that, in Gaussian two-group separation, the optimally discriminating projection direction can be estimated without any knowledge on the group labels. In this presentation, we (a) motivate this estimation problem, and (b) gather several unsupervised estimators based on skewness and derive their limiting distributions. As one of our main results, we show that all affine equivariant estimators of the optimal direction have proportional asymptotic covariance matrices, making their comparison straightforward. We use simulations to verify our results and to inspect the finite-sample behaviors of the estimators.
Location: Hopningspunkten.

Tuesday, May 20, 1.15 pm (NEW TIME), 2025. Seminar in Statistics.

Optimizing the Allocation of Trials to Sub-Regions in Crop Variety Testing
Maryna Prus, Biostatistics Unit, University of Hohenheim
Abstract: download
Location: John von Neumann (CHANGE OF ROOM!) .

Tuesday, June 10, 3.15pm, 2025. Seminar in Mathematical Statistics.

Resurrecting pseudo-inverses: Asymptotic properties of large dimensional Moore-Penrose and Ridge-type inverses with applications
Nestor Parolya, Delft University of Technology
Abstract: In this talk, we discuss high-dimensional asymptotic properties of the Moore-Penrose inverse and, as a byproduct, of various ridge-type inverses of the sample covariance matrix. In particular, the analytical expressions of the asymptotic behavior of the weighted sample trace moments of generalized inverse matrices are deduced in terms of the partial exponential Bell polynomials which can be easily computed in practice. The asymptotic results are obtained without assumption of normality and in the high-dimensional asymptotic regime. Our findings provide universal methodology for construction of fully data-driven improved shrinkage estimators of the precision matrix, optimal portfolio weights and beyond. It is found that the Moore-Penrose inverse acts asymptotically as a certain regularizer of the true covariance matrix and it seems that its proper transformation (shrinkage) performs similarly to or even outperforms the existing benchmarks in many applications, while keeping the computational time as minimal as possible.
Location: Hopningspunkten.