\(U\)-max-statistics and limit theorems for perimeters and areas of random polygons (Q2443258)

\textit{W. Lao} and \textit{M. Mayer} [J. Multivariate Anal. 99, No. 9, 2039--2052 (2008; Zbl 1153.62046)] considered the behaviour of \(U\)-max statistics, the maximum observed value obtained by the kernel function of a \(U\)-statistic. Given independent and identically distributed random variables \(\xi_1, \dots, \xi_n\) and a symmetric function \(h:\mathbb R^m \rightarrow \mathbb R\), then the \(U\)-max statistic is max \(h(\xi_{i_1}, \dots, \xi_{i_m})\), where the maximum is over all \({n \choose m}\) \(m\)-tuples with \(1 \leq i_1 < \ldots < i_m \leq n\). In particular, Lao and Mayer [loc. cit.] applied their results to study the largest area and perimeter of an inscribed triangle, where the vertices are independent and identically distributed points on the unit sphere in \(d\) dimensions, \(d \geq 2\). This paper develops the theory to obtain results for the distribution of the maximum area and perimeter of \(m\)-polygons, \(m\geq 3\), with random vertices on the unit circle. Further the behaviour of the minimal perimeter and minimal area of a circumscribed \(m\)-polygon are also established.