Bootstrapping point processes with some applications (Q1613654)

The authors study the weak limit behavior of point processes obtained by replacing the original observations by the bootstrap sample. In some cases where the usual bootstrap fails, the authors show that subsampling at the rate \(m(n) = o(n) \rightarrow \infty \) depending on the sample size \(n\rightarrow \infty \) the weak limit exists. If \(m(n)\log \log n/n\rightarrow 0\), then the asymptotic results are valid almost surely. The method is applied to bootstrapping extremes.