Article ID Journal Published Year Pages File Type
5129747 Statistics & Probability Letters 2017 9 Pages PDF
Abstract

In this paper, designs, insensitive to the presence of an outlier for estimating a complete set of orthonormal treatment contrasts representing the factorial effects, are obtained for both complete and fractional factorial designs. Fractional factorials incorporated in blocks are also considered in this study. In case of blocked complete factorials, semi-regular group divisible designs and for fractional factorials, universally optimal plans are found to be robust.

Related Topics
Physical Sciences and Engineering Mathematics Statistics and Probability
Authors
, , ,