1: \begin{abstract}

2: We analyze how the choice of the sampling weight affects the efficiency of the

3: Monte Carlo evaluation of classical time autocorrelation functions. Assuming

4: uncorrelated sampling or sampling with constant correlation length, we propose

5: a sampling weight for which the number of trajectories needed for convergence

6: is independent of the correlated quantity, dimensionality, dynamics, and

7: phase-space density. In contrast, it is shown that the computational cost of

8: the \textquotedblleft standard\textquotedblright\ intuitive algorithm which

9: samples directly from the phase-space density may scale exponentially with the

10: number of degrees of freedom. Yet, for the stationary Gaussian distribution of

11: harmonic systems and for the autocorrelation function of a linear function of

12: phase-space coordinates, the computational cost of this standard algorithm is

13: also independent of dimensionality.

14: 

15: \end{abstract}