New procedures for selection among (simulated) alternatives
WSC '77 Proceedings of the 9th conference on Winter simulation - Volume 1
Design and analysis of simulation experiments
WSC '78 Proceedings of the 10th conference on Winter simulation - Volume 1
Hi-index | 0.00 |
In many simulation studies there are available several alternatives and the experimenter is simulating in order to choose which one is best with regard to certain specified criteria. Once an alternative has been selected, people may raise the question: “How good is it?” For this problem one naturally tries to provide a confidence interval for the parameter of the best alternative. Estimation of the best alternative has been studied in the last decade, and most of the works are in the category of k-population problems under one factor. Below we summarize these results and extend them to experiments in factorial designs - one-way and two-way classifications.