The use of antithetic control variates in computer simulations
WSC '81 Proceedings of the 13th conference on Winter simulation - Volume 1
Designing simulation experiments for evaluating manufacturing systems
WSC '94 Proceedings of the 26th conference on Winter simulation
A decomposition approach to variance reduction
WSC '85 Proceedings of the 17th conference on Winter simulation
Design of simulation experiments with manufacturing applications
WSC '93 Proceedings of the 25th conference on Winter simulation
WSC '86 Proceedings of the 18th conference on Winter simulation
| Hi-index | 0.00 |
A number of sampling schemes are described which aim to improve the accuracy of estimators in simulation experiments. The schemes negatively correlate key sample statistics in pairs of runs or blocks of runs. Use is made of generators which produce random samples from certain distributions when sample statistics like the mean and dispersion are prescribed. Special cases considered include normal, inverse Gaussian and gamma generators. These can be used in their own right or as the basis of generators of other distributions. Guide lines are given which indicate the conditions under which the schemes might be effective.