Summary
Williams-designs are proved to be optimal within the class of latin 4×4-squares in rowcolumn-models, where serial correlations are fitted with an autoregressive scheme of first order.
Designs are presented that are not generalized Youden-designs and that are better than Williams-designs for strong negative correlations. However, it will be shown that optimal designs do not depend on the special structure of the error covariance in models with only two columns and in models with a completely symmetric covariance matrix.
Similar content being viewed by others
Literaturverzeichnis
Berenblut, I.I., undG.I. Webb: Experimental design in the presence of autocorrelated errors. Biometrika61, 1974, 427–437.
Budde, M.:D-optimale Versuchspläne für einfache lineare Modelle mit korrelierten Beobachtungen. Diplomarbeit, Universität Dortmund 1981.
Cheng, C.-S., undC.-F. Wu: Balanced repeated measurements designs. Ann. Statist.8, 1980, 1272–1283.
Feld, M.: D-optimale Zeilen-Spaltenpläne für den Vergleich dreier Behandlungen. Diplomarbeit, Universität Dortmund 1981.
Kiefer, J.: Construction and optimality of generalized Youden designs: In: A Survey of Stat. Designs and Linear Models. Amsterdam 1975.
Kiefer, J., undH.P. Wynn: Optimum balanced block and latin square designs for correlated observations. Ann. Statist.9, 1981, 737–757.
Krafft, O.: Lineare statistische Modelle und optimale Versuchspläne. Göttingen 1978.
Kunert, J.: Optimal design and refinement of the linear model with applications to repeated measurements design. Forschungsbericht 81/2, Universität Dortmund, 1981a.
Kunert, J.: Verallgemeinerungen von Sätzen von Cheng und Wu für Modelle mit Nachwirkungseffekten. Forschungsbericht 81/12, Universität Dortmund 1981b.
Sonnemann, E.: D-optimality of complete latin squares. Erscheint in Math. Operationsforsch. Statistik, Ser. Statistics, 1978.
Sonnemann, E.: U-optimale Zeilen-Spaltenpläne für den Vergleich zweier Behandlungen. Erscheint in Metrika, 1979.
Williams, E.J.: Experimental designs balanced for the estimation of residual effects of treatments. Austr.J.Sci.Res.A.2, 1949, 149–168.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Budde, M. Optimale Zweifachblockpläne bei Seriell Korrelierten Fehlern. Metrika 31, 203–213 (1984). https://doi.org/10.1007/BF01915201
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/BF01915201