Nonparametric Smoothing of Directional and Axial Data

Explore nonparametric smoothing techniques for directional and axial data in this 51-minute seminar lecture delivered by Professor Lutz Duembgen from Universität Bern. Learn advanced statistical methods for analyzing data that exists on circular or spherical manifolds, where traditional Euclidean smoothing approaches may not be appropriate. Discover how to handle the unique challenges posed by directional data, such as angles and orientations, and axial data, where opposite directions are considered equivalent. Examine theoretical foundations and practical applications of smoothing techniques specifically designed for these types of data structures. Gain insights into the mathematical frameworks that enable effective nonparametric estimation when dealing with data that has inherent directional or axial properties, and understand how these methods contribute to the broader field of prediction uncertainty in statistics and machine learning.