Change Point Detection for Functional Autoregressive Processes on the Sphere

TL;DR

This paper presents a new method for detecting change points in spherical functional autoregressive processes, using a LASSO-regularized estimator with theoretical guarantees, applicable to climate science and cosmology.

Contribution

It introduces a novel change point detection framework for spherical processes using harmonic domain penalized dynamic programming, without prior knowledge of change points.

Findings

Provides non-asymptotic theoretical guarantees.

Operates without prior knowledge of change points.

Applicable to climate science and cosmology data.

Abstract

We introduce a novel framework for change point detection in spherical functional autoregressive (SPHAR) processes, enabling the identification of structural breaks in spatio-temporal random fields on the sphere. Our LASSO-regularized estimator, based on penalized dynamic programming in the harmonic domain, operates without knowledge of the number or locations of change points and offers non-asymptotic theoretical guarantees. This approach provides a new tool for analyzing nonstationary phenomena on the sphere, relevant to climate science, cosmology, and beyond.