Decorrelation of samples in Quantum Monte Carlo calculations and scaling of autocorrelation time in Li and H$_{2}$O clusters

Abstract

We have investigated decorrelation of samples in Quantum Monte Carlo (QMC) ground-state energy calculations for large Li and H$_{2}$O nanoclusters. Binning data as a way of eliminating statistical correlations, as is the common practice, is found to become increasingly impractical as the system size grows. We demonstrate nevertheless that it is possible to perform accurate energy calculations - without decorrelating samples - by exploiting the scaling of the integrated autocorrelation time $τ$ as a function of the number of electrons in the system.

4 pages, 2 figures, to appear in Phys. Rev. B