Key points are not available for this paper at this time.
We demonstrate that the quantum Monte Carlo (QMC) methodology (i) provides barrier heights and heats of formation within 0. 05eV of experimental values, (ii) confirms recent measurements for the ring inversion of cyclo-octatetraene, and (iii) enables us to predict quantities not yet measured. Density functional methods show a mixed performance in achieving the accuracy required for predictive calculations. Further comparisons show that QMC is competitive in accuracy with the best correlated wave function methods while being applicable to much larger systems because of more favorable scaling.
Grossman et al. (Mon,) studied this question.