The meta-elliptical distributions with given marginals (Q697465)

Let \(\xi_1,\xi_2,\dots\) be independent random variables with the common distribution \(F\) such that \(F((-\infty,0]) <1\), and \({\mathbf E}\xi_1\) exists and equals \(-a<0\). Put \(S_0=0\), \(S_n=\xi_1+ \cdots+\xi_n\) and \(M_n=\max \{S_k,\;0\leq k\leq n\}\). The main result is as follows. If the distribution of \(\xi_1 I\{\xi_1\geq 0\}\) is strongly subexponential, then \[ P(M_n\geq x)= \bigl(1+ \varepsilon_n(x) \bigr)a^{-1} \int_x^{x+na} F\bigl((u,\infty) \bigr) du, \] where \(\varepsilon_n(x)\to 0\) as \(x\to \infty\) uniformly in \(n\geq 1\).