Invariance principle for variable speed random walks on trees

DOI10.1214/15-AOP1071MaRDI QIDQ2412657zbMATH OpenFDO

Authors Wolfgang Löhr, Anita Winter, Siva R. Athreya

Publication date 24 October 2017

Published in The Annals of Probability (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1404.6290, https://projecteuclid.org/euclid.aop/1490947305

zbMATH Keywords

\(\mathbb{R}\)-treeBrownian motionconvergence of Markov chainsdiffusions on metric measure treesDirichlet formGromov-Hausdorff-vague topologyspeed measure

Mathematics Subject Classification ID

Diffusion processes and stochastic analysis on manifolds (58J65) Probability measures on topological spaces (60B05) Probability theory on algebraic and topological structures (60B99) Discrete-time Markov processes on general state spaces (60J05) Continuous-time Markov processes on general state spaces (60J25) Continuous-time Markov processes on discrete state spaces (60J27) Branching processes (Galton-Watson, birth-and-death, etc.) (60J80)

Abstract: We consider stochastic processes on complete, locally compact tree-like metric spaces $(T,r)$ on their "natural scale" with boundedly finite speed measure $u$. Given a triple $(T,r,u)$ such a speed-$u$ motion on $(T,r)$ can be characterized as the unique strong Markov process which if restricted to compact subtrees satisfies for all $x,yinT$ and all positive, bounded measurable $f$, [ mathbb{E}^x [ int^{ au_y}_0mathrm{d}s, f(X_s) ] = 2int_T u(mathrm{d}z), r(y,c(x,y,z))f(z) < infty, ] where $c(x,y,z)$ denotes the branch point generated by $x,y,z$. If $(T,r)$ is a discrete tree, $X$ is a continuous time nearest neighbor random walk which jumps from $v$ to $v^{\prime }simv$ at rate $frac12cdot(u(v)cdotr(v,v^{\prime }){)}^{-1}$. If $(T,r)$ is path-connected, $X$ has continuous paths and equals the $u$-Brownian motion which was recently constructed in [AthreyaEckhoffWinter2013]. In this paper we show that speed-$u_n$ motions on $(T_n,r_n)$ converge weakly in path space to the speed-$u$ motion on $(T,r)$ provided that the underlying triples of metric measure spaces converge in the Gromov-Hausdorff-vague topology introduced recently in [AthreyaLohrWinter2016].

Review text

From the authors' introduction: Some Markov processes ``have in common that their state spaces are closed subsets of the real line and that their random trajectories `do not jump over points'. When put in their `natural scale' these processes are determined by their `speed measure'. The most classical example is the symmetric simple random walk on \(\mathbb{Z}\) which, after a suitable rescaling, converges to standard Brownian motion. You ``might think of the rescaled random walk as such a process with speed measure \(\frac{1}{{\sqrt n }}q(\sqrt n \bullet )\), where \(q\) denotes the counting measure on \(\mathbb{Z}\), and of the standard Brownian motion as such a process whose speed measure equals the Lebesgue measure on \(\mathbb{R}\). We want to extend this result ``to Markov process which take values in tree-like metric spaces. Before we state our main result precisely, we do the preliminary work and define the space of rooted metric boundedly finite measure trees equipped with pointed Gromov-vague topology and give our notion of convergence in path space. Denote by \(\mathbb{T}\) the set of all equivalent classes of rooted metric boundedly finite measure trees. ``Our main result is the following. Theorem 1 (Invariance principle). Assume that \(\chi = (T,r,\rho ,\nu ),\) \({\chi _1} = ({T_1},{r_1},{\rho _1},{\nu _1}),\) \({\chi _2} = ({T_2},{r_2},{\rho _2},{\nu _2})\) are in \(\mathbb{T}\). Let \(X\) be the speed-\(\nu\) motion on \((T, r)\), starting in\(\rho,\) and for all \(n\), let \({X^n}\) be the speed-\({\nu _n}\) motion on \(({T_n},{r_n})\) started in \({\rho _n}\). Assume that the following conditions hold: {\parindent=10mm \begin{itemize}\item[(A0)] For all \(R>0\), \(\limsup_{n \to \infty} \sup \{ {r_n}(x,z):x \in {B_n}({\rho _n},R),z \in {T_n},x \sim z\} < \infty \). \item[(A1)] The sequence \({({\chi _n})_{n \in N}}\) converges to \(\chi \) pointed Gromov-vaguely. \item[(A2)] The uniform local lower mass-bound property holds. \end{itemize}} Then the following holds: {\parindent=6mm \begin{itemize}\item[(i)] \({X^n}\) converges in the one-point compactification weakly in path-space to a process \(Y\), such that \(Y\) stopped at infinity has the same distribution as the speed-\(\nu \) motion \(X\). In particular, if \(X\) is conservative (i.e., does not hit infinity), then \({X^n}\) converges weakly in path-space to \(X\). \item[(ii)] If \({\sup _{n \in N}}\text{{diameter}}({T_n},{r_n}) < \infty \), and assuming (A1) but not (A2), then \({X^n}\) converges f.d.d. to \(X\). \end{itemize}}

Recommendations

• Invariance principles for random walks in random environment on trees
• On the speed of random walks on random trees
• scientific article; zbMATH DE number 747031
• On conditional invariance principle for random walks
• Invariance principle for non-homogeneous random walks
• scientific article; zbMATH DE number 1984426
• Random walks on trees and an inequality of means
• The speed of random walk on Galton-Watson trees with vanishing conductances
• Differentiability of the speed of biased random walks on Galton-Watson trees

Cited in

(33)

• Scaling limits of stochastic processes associated with resistance forms
• Ricci-Ollivier curvature of the rooted phylogenetic subtree-prune-regraft graph
• Convergence of continuous stochastic processes on compact metric spaces converging in the Lipschitz distance
• The random conductance model with heavy tails on nested fractal graphs
• A functional limit theorem for coin tossing Markov chains
• Stochastic PDEs on graphs as scaling limits of discrete interacting systems
• Quenched and averaged tails of the heat kernel of the two-dimensional uniform spanning tree
• A geometric representation of fragmentation processes on stable trees
• Scaling limits of the three-dimensional uniform spanning tree and associated random walk
• Random generation and scaling limits of fixed genus factorizations into transpositions
• Random walk on barely supercritical branching random walk
• The Aldous chain on cladograms in the diffusion limit
• Invariance principles for random walks in random environment on trees
• Convergence of Brownian motions on metric measure spaces under Riemannian curvature-dimension conditions
• Scaling limit for the ant in a simple high-dimensional labyrinth
• The Kingman tree length process has infinite quadratic variation
• Scaling limit of linearly edge-reinforced random walks on critical Galton-Watson trees
• Scaling limit for random walk on the range of random walk in four dimensions
• Brownian motion on \(\mathbb R\)-trees
• scientific article; zbMATH DE number 1984426 (Why is no real title available?)
• Reversal property of the Brownian tree
• Spaces of algebraic measure trees and triangulations of the circle
• Exact simulation of the genealogical tree for a stationary branching population and application to the asymptotics of its total length
• A combinatorial representation for the invariant measure of diffusion processes on metric graphs
• Mini-workshop: New horizons in motions in random media. Abstracts from the mini-workshop held February 26 -- March 4, 2023
• A unified framework for generalizing the Gromov-Hausdorff metric
• Anomalous scaling regime for one-dimensional Mott variable-range hopping
• Backbone scaling limits for random walks on random critical trees
• Algebraic two-level measure trees
• Aging and sub-aging for one-dimensional random walks amongst random conductances
• The gap between Gromov-Vague and Gromov-Hausdorff-vague topology
• Dynamics on accelerated growth tree networks: random walks and consensus problem
• Local limit theorems for sequences of simple random walks on graphs

This page was built for publication: Invariance principle for variable speed random walks on trees

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2412657)