Central limit theorems for random polygons in an arbitrary convex set (Q533742): Difference between revisions

Consider a Poisson point process in a convex set \(K\subset {\mathbb R}^2\) of intensity equal to the Lebesgue measure. Denote by \(\Pi_K\) the convex hull of the points of this process. \(\Pi_K\) is called a random Poisson polygon. Let \(N=N(\Pi_K)\) be the number of vertices of \(\Pi_K\) and \(A=A(\Pi_K)\) be the area of \(K\setminus \Pi_K\). In the paper the author studies the probability distributions of these random variables. The main result is a central limit theorem for both the area and the number of vertices (Theorem 1.1), which is uniform over the set of all convex \(K\). Note that the author requires no regularity on \(\partial K\). Previous results on random polygons analogous to Theorem 1.1 have been confined to two cases: 1) \(K\) is a polygon and 2) \(\partial K\) is of class \(C^2\) with nonvanishing curvature. In recent years there has been significant progress in the study of random polytopes, but again with additional conditions 1) and 2), see [\textit{I. Bárány} and \textit{M. Reitzner}, Ann. Probab. 38, No.~4, 1507--1531 (2010; Zbl 1204.60018); and Adv. Math. 225, No.~4, 1986--2001 (2010; Zbl 1204.52007)].