Compound Poisson statistics in conventional and nonconventional setups (1404.3516v2)
Abstract: Given a periodic point $\omega$ in a $\psi$-mixing shift with countable alphabet, the sequence ${S_{n}}$ of random variables counting the number of multiple returns to shrinking cylindrical neighborhoods of $\omega$ is considered. Necessary and sufficient conditions for the convergence in distribution of ${S_{n}}$ are obtained, and it is shown that the limit is a Polya-Aeppli distribution. A global condition on the shift system, which guarantees the convergence in distribution of ${S_{n}}$ for every periodic point, is introduced. This condition is used to derive results for $f$-expansions and Gibbs measures. Results are also obtained concerning the possible limit distribution of sub-sequences ${S_{n_{k}}}$. A family of examples in which there is no convergence is presented. We exhibit also an example for which the limit distribution is pure Poissonian.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Collections
Sign up for free to add this paper to one or more collections.