Uniform Markov renewal theory and ruin probabilities in Markov random walks. (Q1879907): Difference between revisions

A Markov random walk \(((X_n,S_n),n\geq 0)\) with a general state space is considered. It is assumed that the Markov chain \((X_n,n\geq 0)\) has an invariant probability \(\pi\). For \(b\geq 0\), define the stopping time \(\tau=\tau(b)=\inf\{n:S_n>b\},\;\tau_{+}=\tau(0).\) The author studies the asymptotic approximation of the first passage probabilities \(P_{\pi}(\tau< m)\) and \(P_{\pi}(\tau< mS_m<c)\) where \(m\leq \infty\), \(c\leq b\). The asymptotic behavior is defined as \(m\to\infty\), and \(b=\zeta \sqrt{m}\), \(c=\gamma \sqrt{m}\) for some \(\gamma\leq \zeta\), \(\zeta>0\). For related papers see: \textit{S. Grigorescu} and the reviewer [Z. Wahrscheinlichkeitstheorie Verw. Geb. 35, 65--73 (1976; Zbl 0314.60046)]; \textit{N. Limnios} and the reviewer [Rev. Roum. Math. Pures Appl. 44, 75--83 (1999; Zbl 0995.60034)]; \textit{D. S. Silvestrov} [Acta Appl. Math. 34, No. 1--2, 109--124 (1994; Zbl 0802.60078)].