arXivDaily arXiv每日学术速递 周一至周五更新
稍安勿躁,官方尚未更新
arXiv 2402.08548math.PR
收藏

分支区间划分扩散过程

Branching Interval Partition Diffusions

Matthew Buckland

AI总结:

本文引入并研究一般情形下的分支区间划分扩散过程,推广了自相似情形的先前工作,由此产生了一类特征在生命周期内呈扩散性变化的新型Crump-Mode-Jagers型分支过程。

摘要:

我们引入并研究了一般情形下的分支区间划分扩散过程。我们让区间宽度根据一般的实值扩散过程独立演化,仅受限于确保区间有限寿命并允许连续生成新区间的条件。后者由实值扩散过程的Pitman-Yor游程测度及相关的谱正Lévy过程控制,以对这些游程及其起始时间进行排序。这推广了Forman、Pal、Rizzolo和Winkel关于自相似情形的先前工作,并产生了一类新的具有一般马尔可夫齐次性的Crump-Mode-Jagers型分支过程,其特征在生命周期内呈扩散性变化。

英文摘要:

We introduce and study branching interval partition diffusions in their natural generality. We let interval widths evolve independently according to a general real-valued diffusion subject only to conditions that ensure finite lifetimes of intervals and allow the continuous generation of new intervals. The latter is governed by the Pitman-Yor excursion measure of the real-valued diffusion and an associated spectrally positive Lévy process to order both these excursions and their start times. This generalises previous work by Forman, Pal, Rizzolo and Winkel on the self-similar case and gives rise to a new class of general Markovian homogeneous Crump-Mode-Jagers-type branching processes with characteristics varying diffusively during their lifetimes.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑