Simulation methodology for statisticians, operations analysts, and engineers: vol. 1
Simulation methodology for statisticians, operations analysts, and engineers: vol. 1
| Hi-index | 0.00 |
Nonlinear regression-adjusted control variables are investigated for improving variance reduction in statistical and system simulations. Simple control variables are transformed using linear and nonlinear transformations, and parameters of these transformations are selected using linear or nonlinear least squares regression. As an example, piecewise powertransformed variables are used in the estimation of the mean for the two variable Anderson-Darling goodness-of-fit statistic W22. Substantial variance reduction over straightforward controls is obtained. These parametric transformations are compared against optimal, additive, nonparametric transformations from ACE and are shown to be nearly optimal.