search query: @indexterm sampling / total: 296
reference: 87 / 296
| Author: | Devroye, L. |
| Title: | Coupled samples in simulation. |
| Journal: | Operations Research
1990 : JAN-FEB, VOL. 38:1, p. 115-126 |
| Index terms: | SIMULATION SAMPLING |
| Language: | eng |
| Abstract: | It is assumed that two samples of independent identically distributed random variables are wished to be generated, where the variables have different densities. When these samples are used in a simulation, experimental results for varying distributions are presented, and the densities are close, it is sometimes desirable to have close simulation outcomes. This can be achieved by insisting that both samples agree in most of their components, for as many components as possible under the given distributional constrains. Samples with this property are said to be optimally coupled. Various methods of coupling two samples are proposed and studied. |
SCIMA