|
Registro 1 de 2, Base de información BIBCYT |
|
|
Información de existencia
|
Resumen
A maximum estimability (maxest) criterion is proposed for design classification and selection. It is an extension and refinement of Webb's resolution criterion for general factorial designs. By using the estimability vector associated with the maxest criterion, projective properties of nonregular designs are studied from the estimability perspective. Comparisons with other criteria are also discussed.
|
|
Registro 2 de 2, Base de información BIBCYT |
|
|
Información de existencia
|
Resumen
Rechtschaffner designs are saturated designs of resolution V in which main effects and two-factor interactions are estimable if three-factor and higher order interactions are negligible. Statistical properties of Rechtschaffner designs are studied in this paper. Best linear unbiased estimators of main effects and two-factor interactions are given explicitly and asymptotic properties of correlations between these estimators are studied as well. It is shown that designs recommended by Rechtschaffner [1967. Saturated fractions of 2n and 3n factorial designs, Technometrics 9, 569–576] are not only A-optimal but also D-optimal. Comparisons of Rechtschaffner designs with other A- and D-optimal designs of resolution V are also discussed. |