Analysis of Youden square design with several missing observations

Abstract

The need for estimating missing values has long been recognized by workers dealing with experimental design. All the numerous methods that have been advanced by various workers in this field during the last five decades can broadly be classified into three categories: non iterative procedure, iterative procedure, and covariance technique. In the present study, the authors have tried to discuss the analysis of Youden Square Design in presence of several missing observations occurring in any manner whatsoever and singular pattern of missing observations as well by using Bartlett’s covariate technique. The expressions for estimates of missing observations and variance of the various elementary treatment contrasts have been obtained.