Large excursions and conditioned laws for recursive sequences generated by random matrices

DOI10.1214/17-AOP1221MaRDI QIDQ1660628zbMATH OpenFDO

Authors Jeffrey F. Collamore, Sebastian Mentemeier

Publication date 16 August 2018

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

Full work available at URL https://arxiv.org/abs/1608.05175, https://projecteuclid.org/euclid.aop/1528876822

zbMATH Keywords

extreme value theoryfirst passage timeslarge deviationsconditional limit theoremsproducts of random matricesstochastic fixed-point equationsrandom recurrence equationsMarkov chain theory in general state spacenonlinear Markov renewal theory

Mathematics Subject Classification ID

Large deviations (60F10) Extreme value theory; extremal stochastic processes (60G70) Random matrices (probabilistic aspects) (60B20) Markov renewal processes, semi-Markov processes (60K15) Sample path properties (60G17)

Abstract: We determine the large exceedance probabilities and large exceedance paths for the matrix recursive sequence $V_n=M_n{V}_{n-1}+Q_n,:n=1,2,ldots,$ where $M_n$ is an i.i.d. sequence of $dimesd$ random matrices and $Q_n$ is an i.i.d. sequence of random vectors, both with nonnegative entries. Early work on this problem dates to Kesten's (1973) seminal paper, motivated by an application to multi-type branching processes. Other applications arise in financial time series modeling (connected to the study of the GARCH($p,q$) processes) and in physics, and this recursive sequence has also been the focus of extensive work in the recent probability literature. In this work, we characterize the distribution of the first passage time $T_u^A:=infn:V_{n}inuA$, where $A$ is a subset of the nonnegative quadrant in ${mathbbR}^d$, showing that $T_u^A/u^{a}lpha$ converges to an exponential law. In the process, we also revisit and refine Kesten's classical estimate, showing that if $V$ has the stationary distribution of $V_n$, then $mathbbPleft(VinuAight)sim{C}_A{u}^{-alpha}$ as $uoinfty$, providing, most importantly, a new characterization of the constant $C_A$. Finally, we describe the large exceedance paths via two conditioned limit laws. In the first, we show that conditioned on a large exceedance, the process $V_n$ follows an exponentially-shifted Markov random walk, which we identify, thereby generalizing results for classical random walk to matrix recursive sequences. In the second, we characterize the empirical distribution of $log|V_n|-log|V_{n-1}|$ prior to a large exceedance, showing that this distribution converges to the stationary law of the exponentially-shifted Markov random walk.

Recommendations

• Large exceedences for uniformly recurrent Markov-additive processes and strong-mixing stationary processes
• Large deviations for point processes based on stationary sequences with heavy tails
• Large deviation estimates for exceedance times of perpetuity sequences and their dual processes
• Extremal behaviour of models with multivariate random recurrence representation
• Maxima of recurrent random sequences

Cites work

• scientific article; zbMATH DE number 3942642 (Why is no real title available?)
• scientific article; zbMATH DE number 1191612 (Why is no real title available?)
• scientific article; zbMATH DE number 44410 (Why is no real title available?)
• scientific article; zbMATH DE number 3604123 (Why is no real title available?)
• scientific article; zbMATH DE number 3349081 (Why is no real title available?)
• A characterization of multivariate regular variation.
• A note on multitype branching process with bounded immigration in random environment
• A probabilistic representation of constants in Kesten's renewal theorem
• Conditioned limit theorems relating a random walk to its associate, with applications to risk reserve processes and the GI/G/1 queue
• Extremal behaviour of solutions to a stochastic difference equation with applications to ARCH processes
• Extremal behaviour of stationary Markov chains with applications
• Extremal theory for stochastic processes
• Extreme Value Theory for a Class of Markov Chains with Values in ℝ^d
• Extreme Values in the GI/G/1 Queue
• General Irreducible Markov Chains and Non-Negative Operators
• Heavy tail phenomenon and convergence to stable laws for iterated Lipschitz maps
• Implicit renewal theory and tails of solutions of random equations
• Large deviation estimates for exceedance times of perpetuity sequences and their dual processes
• Large deviation tail estimates and related limit laws for stochastic fixed point equations
• Large exceedances for multidimensional Lévy processes
• Large excursions and conditioned laws for recursive sequences generated by random matrices
• Limit theorems for products of positive random matrices
• Markov chains and stochastic stability
• Nonlinear Markov renewal theory with statistical applications
• Nonlinear renewal theory for Markov random walks
• On conditioning a random walk to stay nonnegative
• On generalized multiplicative cascades
• On multidimensional Mandelbrot cascades
• On multivariate stochastic fixed point equations. The smoothing transform and random difference equations
• On the Harris recurrence of iterated random Lipschitz functions and related convergence rate results
• On the Kesten-Goldie constant
• On the Theory of Markov Renewal
• On the foundations of multivariate heavy-tail analysis
• On the stationary tail index of iterated random Lipschitz functions
• One-dimensional linear recursions with Markov-dependent coefficients
• Precise large deviation results for products of random matrices
• Random difference equations and renewal theory for products of random matrices
• Random recurrence equations and ruin in a Markov-dependent stochastic economic environment
• Random walks in a random environment
• Rare event simulation for processes generated via stochastic fixed point equations
• Renewal theory for functionals of a Markov chain with general state space
• Sequential analysis. Tests and confidence intervals
• Spectral gap properties for linear random walks and Pareto's asymptotics for affine stochastic recursions
• Stochastic Models with Power-Law Tails
• Support theorems for the Radon transform and Cramér-Wold theorems
• Sur une extension de la notion de loi semi-stable. (On an extension of the notion of semi-stable law)
• Tail estimates for stochastic fixed point equations via nonlinear renewal theory
• Weak convergence and empirical processes. With applications to statistics

Cited in

(9)

• Large excursions and conditioned laws for recursive sequences generated by random matrices
• Properties of multitype subcritical branching processes in random environment
• The survival probability for a class of multitype subcritical branching processes in random environment
• Heavy tails for an alternative stochastic perpetuity model
• Limit theory and robust evaluation methods for the extremal properties of GARCH\((p,q)\) processes
• Multi-type subcritical branching processes in a random environment
• Sample-path large deviations for a class of heavy-tailed Markov-additive processes
• Large deviation estimates for exceedance times of perpetuity sequences and their dual processes
• Large exceedences for uniformly recurrent Markov-additive processes and strong-mixing stationary processes

This page was built for publication: Large excursions and conditioned laws for recursive sequences generated by random matrices

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1660628)