Variance Reduction Using Nonlinear Control and Transformations
Abstract
Nonlinear regression-adjusted control variables are investigated for improving variance reduction in statistical and system simulations. To this end, simple control variables are piecewise sectioned and then transformed using linear and nonlinear transformations. Optimal parameters of these transformations are selected using linear or nonlinear leastsquares regression algorithms. As an example, piecewise power-transformed variables are used in the estimation of the mean for the two-variable Anderson-Darling goodness-of-fit statistic. Substantial variance reduction over straightforward controls is obtained. These parametric transformations are compared against optimal, additive nonparametric transformations obtained by using the ACE algorithm and are shown, in comparison to the results from ACE, to be nearly optimal.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1988
- Accession Number
- ADA200471
Entities
People
- Peter A. Lewis
- R. Kevin Wood
- Richard L. Ressler
Organizations
- Naval Postgraduate School