Breaking multivariate records

DOI10.1214/23-EJP968MaRDI QIDQ6165981zbMATH OpenOpenAlexFDO

Authors James Allen Fill

Publication date 2 August 2023

Published in Electronic Journal of Probability (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2109.14846

zbMATH Keywords

asymptoticscurrent recordsmaximamultivariate recordsPareto recordsPoisson point processesrecord breaking

Mathematics Subject Classification ID

Geometric probability and stochastic geometry (60D05) Central limit and other weak theorems (60F05) Point processes (e.g., Poisson, Cox, Hawkes processes) (60G55)

Abstract: For a sequence of i.i.d. $d$-dimensional random vectors with independent continuously distributed coordinates, say that the $n$th observation in the sequence sets a record if it is not dominated in every coordinate by an earlier observation; for $jleqn$, say that the $j$th observation is a current record at time $n$ if it has not been dominated in every coordinate by any of the first $n$ observations; and say that the $n$th observation breaks $k$ records if it sets a record and there are $k$ observations that are current records at time $n-1$ but not at time $n$. For general dimension $d$, we identify, with proof, the asymptotic conditional distribution of the number of (Pareto) records broken by an observation given that the observation sets a record. Fix $d$, and let $mathcalK(d)$ be a random variable with this distribution. We show that the (right) tail of $mathcalK(d)$ satisfies [ {mathbb P}({mathcal K}(d) geq k) leq expleft[ - Omega!left( k^{(d - 1) / (d^2 + d - 3)} ight) ight] mbox{as $koinfty$} ] and [ {mathbb P}({mathcal K}(d) geq k) geq expleft[ - O!left( k^{1 / (d - 1)} ight) ight] mbox{as $koinfty$}. ] When $d=2$, the description of $mathcalK(2)$ in terms of a Poisson process agrees with the main result from Fill [Comb. Probab. Comput. 30 (2021) 105--123] that $mathcalK(2)$ has the same distribution as $mathcalG-1$, where $$(1/2)$$. Note that the lower bound on $mathbbP(mathcalK(d)geqk)$ implies that the distribution of $mathcalK(d)$ is NOT (shifted) Geometric for any $dgeq3$. We show that $mathbbP(mathcalK(d)geq1)=exp[-Theta(d)]$ as $doinfty$; in particular, $mathcalK(d)o0$ in probability as $doinfty$.

Recommendations

• Breaking bivariate records
• scientific article; zbMATH DE number 4100426
• Estimates of some probabilities in multidimensional convex records
• On the relationships between the extreme values of two-dimensional random vectors
• On multivariate records from random vectors with independent components

Cites work

• Breaking bivariate records
• scientific article; zbMATH DE number 5988004 (Why is no real title available?)
• Lectures on the Poisson Process
• Random k-dimensional orders: Width and number of linear extensions
• Random orders
• The Pareto record frontier

Cited in

(14)

• On the relationships between the extreme values of two-dimensional random vectors
• Many multivariate records
• The chain records
• scientific article; zbMATH DE number 4100426 (Why is no real title available?)
• On multivariate records from random vectors with independent components
• Estimates of some probabilities in multidimensional convex records
• Breaking bivariate records
• A new type of multivariate records: depth-based records
• On multidimensional record patterns
• Multivariate records based on dominance
• On multivariate extremal processes
• On the probability of a Pareto record
• Multivariate records and hitting scenarios
• Joint space-time modelling for upper daily maximum and minimum temperature record-breaking

This page was built for publication: Breaking multivariate records

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6165981)