Detecting a late changepoint in the preferential attachment model

TL;DR

This paper develops statistical tests to detect recent changes in the attachment mechanism of a preferential attachment network from a single snapshot, focusing on the challenging regime where the changepoint occurs close to the observation time.

Contribution

It introduces two asymptotically powerful tests for late changepoints in preferential attachment models, one requiring knowledge of the initial parameter and one that does not.

Findings

Tests are asymptotically normal, enabling precise calibration.

Powerful when the changepoint is within a certain late regime (γ > 1/2).

Numerical experiments confirm finite sample effectiveness.

Abstract

Motivated by the problem of detecting a change in the evolution of a network, we consider the preferential attachment random graph model with a time-dependent attachment function. Our goal is to detect whether the attachment mechanism changed over time, based on a single snapshot of the network and without directly observable information about the dynamics. We cast this question as a hypothesis testing problem, where the null hypothesis is a preferential attachment model with a constant affine attachment parameter $\delta_0$, and the alternative hypothesis is a preferential attachment model where the affine attachment parameter changes from $\delta_0$ to $\delta_1$ at an unknown changepoint time $\tau_n$. For our analysis we focus on the regime where $\delta_0$ and $\delta_1$ are fixed, and the changepoint occurs close to the observation time of the network (i.e., $\tau_n = n - c n^\gamma$ with $c>0$ and $\gamma \in (0, 1)$). This corresponds to the relevant scenario where we aim to detect the changepoint shortly after it has happened. We present two tests based on the number of vertices with minimal degree, and show that these are asymptotically powerful when $\tfrac{1}{2}<\gamma<1$. We conjecture that there is no powerful test based on the final network snapshot when $\gamma < \tfrac{1}{2}$. The first test we propose requires knowledge of $\delta_0$. The second test is significantly more involved, and does not require the knowledge of $\delta_0$ while still achieving the same performance guarantees. Furthermore, we prove that the test statistics for both tests are asymptotically normal, allowing for accurate calibration of the tests. This is demonstrated by numerical experiments, that also illustrate the finite sample test properties.