Optimal block designs for three treatments when observations are correlated
Abbreviated Journal Title
J. Stat. Plan. Infer.
block design; E- and M-optimality; information matrix; universal; optimality; ERRORS; Statistics & Probability
E-optimal designs forcomparing three treatments in blocks of sizethree are identified, where intrablock observations are correlated according to a first order autoregressive error process with parameter rho is an element of (0, 1). For number of blocks b of the form b = 3n + 1, there are two distinct optimal designs depending on the value of rho, with the best design being unequally replicated for large rho. For other values of b, binary, equireplicate designs with specified within-block assignment patterns are best. In many cases, the stronger majorization optimality is established. (C) 2007 Elsevier B.V. All rights reserved.
Journal of Statistical Planning and Inference
"Optimal block designs for three treatments when observations are correlated" (2008). Faculty Bibliography 2000s. 1075.