OPTIMAL ENERGY ESTIMATION IN PATH-INTEGRAL MONTE-CARLO SIMULATIONS

We investigate the properties of two standard energy estimators used i n path-integral Monte Carlo simulations. By disentangling the variance of the estimators and their autocorrelation times we analyze the depe ndence of the performance on the update algorithm and present a detail ed comparison of refined update schemes such as multigrid and staging techniques, We show that a proper combination of the two estimators le ads to a further reduction of the statistical error of the estimated e nergy with respect to the better of the two without extra cost. (C) 19 97 American Institute of Physics.