Improving the accuracy of estimators for the two-point correlation function

We show how to increase the accuracy of estimates two-point correlation function without sacrificing efficiency. quantify error pair-counts and Landy-Szalay estimator by comparing them with exact reference values. The standard method, using random point sets, is compared geometrically motivated estimators quasi-Monte~Carlo integration. In scales proportionally $1/\sqrt{N_r}$, $N_r$ being number points. our improved methods, almost $1/N_q$, where $N_q$ points from a low-discrepancy sequence. study run times new in comparison those estimator, keeping same level accuracy. For considered case, we always see speedup ranging 50% up factor several thousand. also discuss apply these incompletely sampled galaxy catalogues.