Source author record

Mariusz Maślanka

Explore this author’s publications, collaborators and research topics. This catalog record has not been claimed by its author.

ResearcherUnclaimed source record

Catalog footprint

Research overview

2works
1topics
1close collaborators

Published work

2 publications

preprint2016arXiv

Precise large deviations of the first passage time

Let SnS_n be partial sums of an i.i.d. sequence {Xi}\{X_i\}. We assume that EX1<0\mathbb{E} X_1 <0 and P[X1>0]>0\mathbb{P}[X_1>0]>0. In this paper we study the first passage time τu=inf⁡{n:  Sn>u}. τ_u = \inf\{n:\; S_n > u\}. The classical Cramér's estimate of the ruin probability says that P[τu<∞]∼Ce−α0uas u→∞, \mathbb{P}[τ_u<\infty] \sim C e^{-α_0 u}\quad \text{as } u\to \infty, for some parameter α0α_0. The aim of the paper is to describe precise large deviations of the first crossing by SnS_n a linear boundary, more precisely for a fixed parameter ρρ we study asymptotic behavior of P[τu=⌊u/ρ⌋]\mathbb{P}\left[τ_u = \lfloor u/ρ\rfloor \right] as uu tends to infinity.

preprint2016arXiv

Tail asymptotics of maximums on trees in the critical case

We consider solutions to the maximum recursion on weighted branching trees given by$$X\,{\buildrel d\over=}\,\bigvee_{i=1}^{N}{A_iX_i}\vee B,$$where NN is a random natural number, BB and {Ai}i∈N\{A_i\}_{i\in\mathbb{N}} are random positive numbers and XiX_i are independent copies of XX, also independent of NN, BB, {Ai}i∈N\{A_i\}_{i\in\mathbb{N}}. Properties of solutions to this equation are governed mainly by the function m(s)=E[∑i=1NAis]m(s)=\mathbb{E}\big[\sum_{i=1}^NA_i^s\big]. Recently, Jelenković and Olvera-Cravioto proved, assuming e.g. m(s)<1m(s)<1 for some ss, that the asymptotic behavior of the endogenous solution RR to the above equation is power-law, i.e.P[R>t]∼Ct−α\mathbb{P}[R>t]\sim Ct^{-α}for some α>0α>0 and C>0C>0. In this paper we assume m(s)≥1m(s)\ge 1 for all ss and prove analogous results.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.