Ordinal patterns in clusters of subsequent extremes of regularly varying time series (Q2027087): Difference between revisions

Let \(\{X_t\}_{t\in\mathbb{Z}}\) be a regularly varying, strictly stationary, \(\alpha\)-mixing time series such that \[ \sup_{A,B\in\mathcal{B}(\mathbb{R}^\mathbb{N})} \left|\mathbb{P}\left(\{X_t\}_{t\leqslant 0}\in A, \{X_t\}_{t\geqslant n}\in B \right)-\mathbb{P}\left(\{X_t\}_{t\leqslant 0}\in A\right)\mathbb{P}\left(\{X_t\}_{t\geqslant n}\in B\right)\right|\\ =O\left(n^{-\delta}\right) \] for some \(\delta>1\). Under suitable additional conditions, the authors consider asymptotical properties of the ratio estimator \[ \widehat{R}_{n,u_n}(A,A_0)=\frac{\widehat{P}_{n,u_n}(A)}{\widehat{P}_{n,u_n}(A_0)}, \] where \[A,A_0\subset [0,\infty)\times(1,\infty)\times[0,\infty)^{t},\] \[ \widehat{P}_{n,u_n}(A)=\frac{1}{n}\sum_{k=1}^{n-t}\mathbb{I}_{\{\{X_i\}_{i=k-1}^{k+t}\in uA\}}, \] and \(u_n\) is some unboundedly increasing sequence such that \(n\mathbb{P}(X_0>u_n)\rightarrow\infty\).