Invariant measures of critical spatial branching processes in high dimensions

成果类型:
Article
署名作者:
Bramson, M; Cox, JT; Greven, A
署名单位:
University of Wisconsin System; University of Wisconsin Madison; Syracuse University; University of Erlangen Nuremberg
刊物名称:
ANNALS OF PROBABILITY
ISSN/ISSBN:
0091-1798
发表日期:
1997
页码:
56-70
关键词:
Diffusions SYSTEM time
摘要:
We consider two critical spatial branching processes on R-d: critical branching Brownian motion, and the critical Dawson-Watanabe process. A basic feature of these processes is that their ergodic behavior is highly dimension dependent. It is known that in low dimensions, d less than or equal to 2, the only invariant measure is delta(0), the unit point mass on the empty state. In high dimensions, d greater than or equal to 3, there is a family {nu(0), theta is an element of [0, infinity)} of extremal invariant measures; the measures nu(0) are translation invariant and indexed by spatial intensity. We prove here, for d greater than or equal to 3, that all invariant measures are convex combinations of these measures.