An effective statistic and robust block designs for studying outliers in incomplete multiresponse experiment

An Andrews-Pregibon statistic has been developed to test outliers in the linear model for incomplete multiresponse experiment designs. An incomplete multiresponse design's data set is used to test this newly developed statistic. A new criterion for the robustness of incomplete multiresponse experimental designs against the presence of a single outlier is proposed. The average Cook-statistic is the basis for this new criterion. A number of one-way elimination of heterogeneity designs are examined that are robust toward the presence of a single outlier.