Multidimensional multiscale scanning in Exponential Families: Limit theory and statistical consequences

TL;DR

This paper develops a unified multiscale scanning methodology for anomaly detection in high-dimensional exponential family data, providing Gaussian approximations, limit theorems, and optimality results.

Contribution

It introduces a new multiscale likelihood ratio testing framework with explicit error control and asymptotic analysis for non-i.i.d. data in exponential families.

Findings

Provides a Gaussian approximation with explicit convergence rates.

Establishes a weak limit theorem as a generalized invariance principle.

Demonstrates minimax optimality for Gaussian cases.

Abstract

We consider the problem of finding anomalies in a $d$-dimensional field of independent random variables $\{Y_i\}_{i \in \left\{1,...,n\right\}^d}$, each distributed according to a one-dimensional natural exponential family $\mathcal F = \left\{F_\theta\right\}_{\theta \in\Theta}$. Given some baseline parameter $\theta_0 \in\Theta$, the field is scanned using local likelihood ratio tests to detect from a (large) given system of regions $\mathcal{R}$ those regions $R \subset \left\{1,...,n\right\}^d$ with $\theta_i \neq \theta_0$ for some $i \in R$. We provide a unified methodology which controls the overall family wise error (FWER) to make a wrong detection at a given error rate. Fundamental to our method is a Gaussian approximation of the distribution of the underlying multiscale test statistic with explicit rate of convergence. From this, we obtain a weak limit theorem which can be seen as a generalized weak invariance principle to non identically distributed data and is of independent interest. Furthermore, we give an asymptotic expansion of the procedures power, which yields minimax optimality in case of Gaussian observations.