A new class of estimators for the reduced second moment measure of point processes

A class of edge-corrected estimators for the reduced second moment measure of a stationary isotropic point process is proposed. All estimators in this class have the same unbiasedness properties as previously suggested edge-corrected estimators. One natural element in this class has superior asymptotic properties to existing estimators when the underlying process is Poisson. A simulation study shows that this superiority also holds for certain cluster processes.