directional-statistics

Non-stationary wrapped Gaussian spatial response model

Isa Marques, Thomas Kneib and Nadja Klein / 2023-11-22

Circular data, i.e., data defined on the unit circle, can be found in many areas of science. The unique nature of these data means that conventional methods for non-circular data are not valid for these. At the same time, advances in geographical information and global positioning systems have…

network-analysis

Populations of Unlabeled Networks: Graph Space Geometry and Generalized Geodesic Principal Components

Anna Calissano, Aasa Feragen and Simone Vantini / 2023-11-06

Sets of graphs arise in many different applications, from medicine to finance, from urban planning to social science. Analysing sets of graphs is far from trivial as they are strongly non Euclidean data type. There are two main sets of graphs: […] Unlabelled graphs poses high challenges from…

webinars

Stochastic Fluid Dynamics

2023-10-14

Stochastic Fluid Dynamics Wednesday, November 15th, 6:00 PT / 9:00 ET / 15:00 CET The study of fluid dynamics equations with white forcing is a classical topic in SPDEs and ergodic theory. Recently a new wave of interest, with a shift in focus towards transport noise, has risen, due to its…

functional-data-analysis

Locally Sparse Functional Regression

Mauro Bernardi, Antonio Canale and Marco Stefanucci / 2023-10-09

In this post we present a new estimation procedure for functional linear regression useful when the regression surface – or curve – is supposed to be exactly zero within specific regions of its domain. Our approach involves regularization techniques, merging a B-spline representation of the unknown…

bayesian-statistics

Linear-cost unbiased estimator for large crossed random effect models via couplings

Paolo Ceriani and Giacomo Zanella / 2023-09-27

In the following we show how it is possible to obtain parallelizable, unbiased and computationally cheap estimates of Crossed random effects models with a linear cost in the number of datapoints (and paramaters) exploiting couplings. […] CREM model a continuous response variables \(Y\) as…

webinars

Algorithmic Fairness

2023-09-19

Algorithmic Fairness Tuesday, October 3rd, 2023, 7:30 PT / 10:30 ET / 16:30 CET 2nd joint webinar of the IMS New Researchers Group, Young Data Science Researcher Seminar Zürich and the YoungStatS Project. When & Where: […] Speakers: […] Abstract: Multi-calibration is a powerful and…