Rates of multivariate normal approximation for statistics in geometric probability (Q6103976)

The authors use stabilization methods and second-order Poincaré inequalities to obtain quantitative central limit theorems for multivariate functionals of Poisson point processes with large intensities. The functional of interest are expressed as sums of score functions which are exponentially stabilizing and satisfy a suitable moment condition. The rates are given in several metrics between multivariate distributions and are shown to be optimal. The obtained general results are applied to deduce rates of multivariate normal convergence for statistics arising in random graphs and topological data analysis as well as for multivariate statistics used to test the equality of distributions.