Extreme value statistics of positive recurrent centrally biased random walks

Roberto Artuso,Manuele Onofri,Gaia Pozzoli,Mattia Radice
DOI: https://doi.org/10.1088/1742-5468/ac98bd
2022-07-15
Abstract:We consider the extreme value statistics of centrally-biased random walks with asymptotically-zero drift in the ergodic regime. We fully characterize the asymptotic distribution of the maximum for this class of Markov chains lacking translational invariance, with a particular emphasis on the relation between the time scaling of the expected value of the maximum and the stationary distribution of the process.
Probability
What problem does this paper attempt to address?