Large deviation probabilities for random walks with semiexponential distributions (Q5932605)

Let \(X_1,X_2,\dots\) be identically distributed independent random variables with distribution function \(F(t)\), and let \[ S_n=\sum_{j=1}^n X_j,\quad \overline S_n(a)=\max_{k\leq n}(S_k-ka). \] It is assumed that \(\mathbf{E}X_1=0\), \(\mathbf{E}X_{1}^{2}=1\), and \(\mathbf{E}|X_1|^b\) is finite for some \(b>2\). It is also assumed that the tail \(\overline F(t)=\mathbf{P}\{X_j\geq t\}\) admits a representation \(\overline F(t)=e^{-t^\alpha\varepsilon(t)}\), where \(\alpha\in(0,1)\) and \(\varepsilon(t)\) is an at infinity slowly varying function which satisfies certain differentiability conditions. The main results concern the asymptotic behavior of the probabilities \(\mathbf{P}\{S_n>x\}\) and \(\mathbf{P}\{\overline S_n(a)>x\}\). Some asymptotic expansions are also presented for these probabilities.