Amsterdam, Netherlands

Inference in Structured High-Dimensional Models

online course
when 17 January 2023 - 20 January 2023
language English
duration 1 week
credits 2 EC
fee EUR 800

The course offers a general approach to draw inference in many high-dimensional models with various hidden structures. We develop a general framework for model+structure combinations and within this framework study various statistical problems such as estimation problem, posterior contraction problem, uncertainty quantification and structure recovery. The proposed general framework unifies a very broad class of high-dimensional models and structures, interesting and important in their own right.

Course leader

E. N. Belitser is an associate professor at the Department of Mathematics at VU, specializing in mathematical statistics.

Target group

This course is suitable for anybody with a strong interest in mathematical statistics, and preference for mathematical preciseness. Especially PhD candidates and master students at the final year, dealing with high-dimensional models, uncertainty quantification and Bayesian methodology can benefit from this course.

A strong background in mathematical statistics and probability is recommended.

Course aim

By the end of this course, students will:

be familiarized with the new notions of modern high-dimensional statistics and empirical Bayes methodology,
be acquainted with the deceptiveness phenomenon in uncertainty quantification problem,
have learnt and understood a general approach to tackle a big family of combinations: high-dimensional model + structure,
have practiced how to apply the general approach to concrete combination model + structure.

Fee info

EUR 800: -Students, PhD students and employees of VU Amsterdam: €500
-Students and PhD students: €600
-Professionals: €800
Applications received before 15 October €50 receive Early Bird Discount

Scholarships

We offer an Equal Access Scholarship. Deadline 15 November. Find out more via www.vu.nl/winterschool