Generalized Gini's mean difference through distortions and copulas, and related minimizing problems (Q6152243)

Let us consider a family of distortion functions \[ H_\alpha:={h_\alpha: [0, 1]\to [0, 1], \alpha\in I\subset R} \] Denote \(X_\alpha\) the random variable with survival function (s.f.) \(F_{X_\alpha}(x)=h_\alpha(F(x))\), for \(x\in R\), and \[ F_{X,h_\alpha}:={X_\alpha: F_{X_\alpha}(x)=h_\alpha(F(x)),\alpha \in I, x\in R} \] the family of all random variables with distorted s.f.'s from \(X\) through \(h_\alpha\in H_\alpha\) [\textit{J. Navarro}, Introduction to system reliability theory. Cham: Springer (2021; Zbl 1494.90001)]. The Gini's mean difference of a random variable X is defined as \[ \mathrm{GMD}(X)=E|X-X'|=2\int F(x)F(x)dx,\tag{\(\ast\)} \] where \(X\) is a random variable and \(X'\) is an independent copy of \(X\); \(F(x)=P(X\leq x)\) the cumulative distribution function (c.d.f), \(F(x)=1-F(x)\) the survival function (s.f.), for \(x\in R\). \(\ell:=\inf\{x\in R: F(x)>0\}\), \(r=\sup\{x\in R: F(x)>0\}\), which may be finite or infinite. Authors extend the Gini's mean difference in Eq. \((\ast)\) in the cases in which the random variables are not necessarily independent or identically distributed. In particular, for \(X\) and \(X_\alpha\) independent random variables, they define the distorted Gini's mean difference of \(X\) as \[ \eta_{X}(\alpha):=E|X-X_\alpha|=\int(F(x)+h_\alpha(F(x))(1-2F(x)))dx. \] More precisely, given a random variable \(X\) and considered a family of its possible distortions, authors define two new measures of distance between \(X\) and each its distortion. For these distance measures, which are extensions of the Gini's mean difference, conditions are determined for the existence of a minimum, or a maximum, within specific families of distortions, generalizing some results presented in [\textit{P. Ortega-Jiménez} et al., ``A minimizing problem of distances between random variables with proportional reversed hazard rate functions'', in: Building bridges between soft and statistical methodologies for data science. Cham: Springer. 311--318 (2023; \url{doi:10.1007/978-3-031-15509-3_41}); \textit{F. Delbaen} and \textit{C. Majumdar}, Front. Math. Finance 2, No. 2, 141--149 (2023; Zbl 07716477)].