scientific article; zbMATH DE number 1295505
For some rv \( X \) and a sequence of rv's \( \{ X_n \} \) denote the family of conditional distributions given \( X=x \) by \( P_x \). Let \( \{ X^x_n \} \) be the corresponding sequence with distribution determined by \( P_x \). If for a properly chosen sequence \( \{ u_n \} \) of real numbers the conditional point processes of exceedances \[ S_n[X^x_n,u_n](.) = \sum_{i=1}^n 1_{ \{X^x_i>u_n\} } \delta_{i/n}(.) \] converge to a point process \( S^x\;\forall x \in \text{Supp }X \), then the sequence \( S_n[X_n,u_n] \) converges in distribution to a point process the distribution of which is a mixture of the distributions of \( S^x \) regulated by the distribution of \( X \). This main result of the paper is illustrated by examples concerning some max-autoregressive processes.
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4248199)