|
| << vorige | volgende >> |
Tijdschrift beschrijving
Alle jaargangen van het bijbehorende tijdschrift
Alle afleveringen van het bijbehorende jaargang
Alle artikelen van de bijbehorende aflevering
Details van artikel 5 van 19 gevonden artikelen
| Titel: |
A scenario for sequential experimentation |
| Auteur: |
Kleijnen, Jack P.C. |
| Verschenen in: |
Communications in statistics |
| Paginering: | Jaargang 15 (1986) nr. 10 pagina's 3067-3078 |
| Jaar: |
1986 |
| Inhoud: |
This tutorial emphasizes the role of differecnt types of experimental design in a multi-stage investigation. In the initial phase group-screeningo can reveal the really important factors among hundreds of factors. Resooulution III designs are useful immediately after the screening phase, to investigate firstorder effects, provided higher-order effects are unimportants, i.e., validation is necessary. Resulotion IV designs may explain why a first-order model is not valid, i.e., they may yield unbiased estimators of sums of interactions. Resolution V designs yield unbiased estimators of the individual two-factor interactions. They can be easily extended to central composite designs to estimate pure quadratic effects of quantitative factors. Smaller steps are also possible, e.g. one run at a time, for model discrimination and calibration. |
| Uitgever: |
Taylor & Francis |
| Bronbestand: |
Elektronische Wetenschappelijke Tijdschriften |
Details van artikel 5 van 19 gevonden artikelen
| << vorige | volgende >> |