Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.29853

Spatial preferential attachment models: robustness revisited

Peter Gracar·Christian Mönch
View original source ↗Read source document ↗Export metadata ↓

Abstract

We investigate robustness under percolation in the degree-driven spatial preferential attachment model. Consider a profile that is regularly varying with exponent $-δ$, where $δ>1$, and an attachment rule with asymptotic slope $γ\in(0,1)$. Jacob and Mörters (Ann. Probab. 45 (2017)) proved robustness when $$ δ\left(\frac1γ-1\right)<1. $$ We prove the converse in the strict regime: if $$ δ\left(\frac1γ-1\right)>1, $$ then sufficiently strong Bernoulli vertex thinning destroys the giant component. In particular, the strict robustness boundary coincides with that of the age-dependent random connection model.

Explore related subjects

math.PR

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Peter Gracar, Christian Mönch. 2026-10-21. Spatial preferential attachment models: robustness revisited. https://arxiv.org/abs/2610.29853

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

The maximum distance covariance of bounded random vectors

A sharp upper bound is obtained for the distance covariance of bounded random vectors in terms of the diameters of their supports alone. The bound is attained for all pairs of support diameter constraints and hence provides the maximum distance covariance subject to constraints of this kind. A normalized measure of dependence is derived by dividing distance covariance by the sharp bound when it is positive. Several properties of the resulting quotient are proved, including domination by distance correlation, invariance under similarities, and subadditivity under sums of independent pairs. The quotient is shown to be nonincreasing under the addition of independent noise, a property not shared by distance correlation. The usual sample distance covariance inherits the sharp bound from the population quantity and the associated normalized statistic is strongly consistent.

math.PR↗
arXiv2026

Ergodicity for stochastic neural field equations

We investigate the well-posedness and long-time behavior of a general continuum neural field model with Gaussian noise on possibly unbounded domains. In particular, we give conditions for the existence of invariant probability measures by restricting the solution flow to an invariant subspace with a nonlocal metric. Under the assumption of a sufficiently large decay parameter relative to the noise intensity, the norm of the connectivity kernel operator, and the Lipschitz regularity of the activation function, we establish exponential ergodicity and exponential mixing of the associated Markovian Feller semigroup and the uniqueness of the invariant measure.

math.PR↗
arXiv2026

On a class of multiplicative Lindley-type recursions with Markov-modulated dependencies

In this paper, we study a class of Markov-modulated multiplicative Lindley-type recursions of the form $W_{n+1}=[V_nW_n+Y_n(V_n)]^+$, where the additive term may depend on the multiplicative coefficient and on a common finite-state background Markov chain. The main feature of the model is that the multiplicative coefficient is allowed to take negative values, which destroys the usual monotonicity structure of Lindley-type recursions. We consider two cases. In the first one, $V_n$ may take the values $1$, a constant $a\in(0,1)$, or a negative value, according to specific probabilities, while $Y_n(V_n)=S_n-A_{n+1}$ and the relevant Laplace--Stieltjes transforms are assumed to be rational. In the second one, $V_n\in\{1,-1\}$, and the additive term has a more general form depending on $V_n$. For both models, we derive vector-valued functional equations for the stationary transform. For the first model, we obtain a recursive transform representation, while for the second model we derive a finite-dimensional characterization based on root-cancellation conditions. We also provide a recursive procedure for computing stationary moments and discuss the light-tailed asymptotic behaviour. An extension based on the Farlie--Gumbel--Morgenstern copula is also presented. Numerical experiments show how the transform representation can be used to compute stationary moments, tail probabilities and quantiles, and validate the results against Monte Carlo simulation.

math.PR↗