On a transformation between distributions obeying the principle of a single big jump (Q2352853)

Let \(\{X_i: i \geq 1\}\) be a sequence of independent and identically distributed random variables with common distribution function \(F.\) Let \(\mathcal I\) denote the set of all distributions \(F\) with unbounded support contained in \([0, \infty)\) such that for all \(n\geq 2\), \(\lim_{K\rightarrow \infty}\liminf_{x\rightarrow \infty}P(X_{n,1}>x-K|S_n>x)=1\), where \(X_{n,1}=\max\{X_i: 1 \leq i \leq n\}\) and \(S_n=\sum_{i=1}^n X_i\). Such a distribution \(F\) in \(\mathcal I\) is said to obey the principle of a single big jump. \textit{S. Beck} et al. [Bernoulli 21, No. 4, 2457--2483 (2015; Zbl 1362.60007)] introduced the class of distributions \(\mathcal I\) and observed that the class \(\mathcal I\) contains many heavy-tailed and light tailed distributions \(F\) obeying the principle of a single big jump. The authors study here the structure of light-tailed distributions in the class \(\mathcal I.\)