Open Access
February 2018 Limit theorems for affine Markov walks conditioned to stay positive
Ion Grama, Ronan Lauvergnat, Émile Le Page
Ann. Inst. H. Poincaré Probab. Statist. 54(1): 529-568 (February 2018). DOI: 10.1214/16-AIHP814

Abstract

Consider the real Markov walk $S_{n}=X_{1}+\cdots+X_{n}$ with increments $(X_{n})_{n\geq 1}$ defined by a stochastic recursion starting at $X_{0}=x$. For a starting point $y>0$, denote by $\tau_{y}$ the exit time of the process $(y+S_{n})_{n\geq 1}$ from the positive part of the real line. We investigate the asymptotic behaviour of the probability of the event $\tau_{y}\geq n$ and of the conditional law of $y+S_{n}$ given $\tau_{y}\geq n$ as $n\to+\infty$.

On considère une marche Markovienne réelle $S_{n}=X_{1}+\cdots+X_{n}$ dont les accroissements $(X_{n})_{n\geq1}$ sont définis par une récursion stochastique partant de $X_{0}=x$. Pour un point de départ $y>0$, on note par $\tau_{y}$ le temps de sortie du processus $(y+S_{n})_{n\geq1}$ de la partie positive de la droite des réels. On s’intéresse au comportement asymptotique de la probabilité de l’évènement $\tau_{y}\geq n$ ainsi qu’à la loi conditionnelle de $y+S_{n}$ sachant $\tau_{y}\geq n$ quand $n\to+\infty$.

Citation

Download Citation

Ion Grama. Ronan Lauvergnat. Émile Le Page. "Limit theorems for affine Markov walks conditioned to stay positive." Ann. Inst. H. Poincaré Probab. Statist. 54 (1) 529 - 568, February 2018. https://doi.org/10.1214/16-AIHP814

Information

Received: 11 January 2016; Revised: 30 November 2016; Accepted: 9 December 2016; Published: February 2018
First available in Project Euclid: 19 February 2018

zbMATH: 06880064
MathSciNet: MR3765899
Digital Object Identifier: 10.1214/16-AIHP814

Subjects:
Primary: 60G50 , 60J05 , 60J50
Secondary: 60G42 , 60J70

Keywords: Exit time , Harmonic function , Markov chains , stochastic recursion

Rights: Copyright © 2018 Institut Henri Poincaré

Vol.54 • No. 1 • February 2018
Back to Top