On convergence to stationarity of fractional Brownian storage

DOI10.1214/08-AAP578MaRDI QIDQ835064zbMATH OpenOpenAlexFDO

Authors Ilkka Norros, M. R. H. Mandjes, Peter W. Glynn

Publication date 27 August 2009

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

Full work available at URL https://arxiv.org/abs/0908.4472

zbMATH Keywords

convergence to stationarityfractional Brownian motionlarge deviationsstorage process

Mathematics Subject Classification ID

Gaussian processes (60G15) Fractional processes, including fractional Brownian motion (60G22) Inventory, storage, reservoirs (90B05)

Abstract: With $M(t):=su{p}_{sin[0,t]}A(s)-s$ denoting the running maximum of a fractional Brownian motion $A(cdot)$ with negative drift, this paper studies the rate of convergence of $mathbbP(M(t)>x)$ to $mathbbP(M>x)$. We define two metrics that measure the distance between the (complementary) distribution functions $mathbbP(M(t)>cdot)$ and $mathbbP(M>cdot)$. Our main result states that both metrics roughly decay as $exp(-vartheta{t}^{2-2H})$, where $vartheta$ is the decay rate corresponding to the tail distribution of the busy period in an fBm-driven queue, which was computed recently [Stochastic Process. Appl. (2006) 116 1269--1293]. The proofs extensively rely on application of the well-known large deviations theorem for Gaussian processes. We also show that the identified relation between the decay of the convergence metrics and busy-period asymptotics holds in other settings as well, most notably when G"artner--Ellis-type conditions are fulfilled.

Review text

Let \(A=(A(t))_{t\geq 0}\) be a fractional Brownian motion and set \[ M(t):=\sup_{0\leq s\leq t}[A(s)-s]\quad\text{and}\quad M:=\sup_{s>0}[A(s)-s]\;. \] The aim of the present paper is to determine the rate of convergence of \(M(t)\) to \(M\) as \(t\to\infty\). Here the distance between \(M(t)\) and \(M\) is measured by two different metrics, namely by the Kolmogorov--Smirnov distance and by the so-called integral distance. In both cases the authors obtain surprisingly sharp convergence rates of exponential order. The proofs of these results rely on large deviation theorems for Gaussian processes.

Recommendations

• On convergence rate to stationarity of queues with general Gaussian input
• Limit theorem for maximum of the storage process with fractional Brownian motion as input
• Large deviations of a storage process with fractional Brownian motion as input
• On the maximum workload of a queue fed by fractional Brownian motion.
• Large buffer asymptotics for the queue with fractional Brownian input

Cites work

• A note on LDP for supremum of Gaussian processes over infinite horizon
• A storage model with self-similar input
• Conditional limit theorems for queues with Gaussian input, a weak convergence approach
• Conditions for exponential ergodicity and bounds for the decay parameter of a birth-death process
• Ergodic Properties of Continuous-Time Markov Processes and Their Discrete Skeletons
• Extremes of a certain class of Gaussian processes
• Fast simulation of overflow probabilities in a queue with Gaussian input
• scientific article; zbMATH DE number 3174847 (Why is no real title available?)
• scientific article; zbMATH DE number 3962933 (Why is no real title available?)
• scientific article; zbMATH DE number 194664 (Why is no real title available?)
• Large buffer asymptotics for the queue with fractional Brownian input
• Large deviations and overflow probabilities for the general single-server queue, with applications
• Large Deviations for Gaussian Queues
• Large deviations for vector-valued Lévy processes
• Large deviations of infinite intersections of events in Gaussian processes
• Large deviations of the sample mean in general vector spaces
• Logarithmic asymptotics for steady-state tail probabilities in a single-server queue
• On exponential ergodicity and spectral structure for birth-death processes. I
• On exponential ergodicity and spectral structure for birth-death processes. II
• On the Correlation Structure of a Lévy-Driven Queue
• On the Dependence Structure of Gaussian Queues
• Sample path large deviations and intree networks
• Sample-path large deviations for tandem and priority queues with Gaussian inputs
• Stochastic Calculus for Fractional Brownian Motion and Applications
• The Exponential Decay of Markov Transition Probabilities

Cited in

(7)

• Approximation of the maximum of storage process with fractional Brownian motion as input
• Limit theorem for maximum of the storage process with fractional Brownian motion as input
• Fractional Brownian Motion withH< 1/2 as a Limit of Scheduled Traffic
• Open problems in Gaussian fluid queueing theory
• Bounds for expected supremum of fractional Brownian motion with drift
• Extremes of nonstationary Gaussian fluid queues
• The Cramér-Lundberg model with a fluctuating number of clients

This page was built for publication: On convergence to stationarity of fractional Brownian storage

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q835064)