Pooled analysis of dependent sets of data

Abstract

A new methodology for the analysis of data generated from experiments in which observations constitute repeated measurements from the same experimental unit at different points of time was developed. The problem of dependence of error terms in successive observations was taken care of in the model for analysis itself. The model included regression of error terms on those in the yester years/seasons. The error mean square from this model was derived using principle of least squares. The proposed method was compared with the widely adopted split-plot analysis and its superiority discussed. The method was illustrated using data generated from an experiment conducted to compare three varieties of alfalfa laid out in RBD with six replications and observations taken in four consecutive seasons. The superiority of the new method over the split-plot analysis was evident in the example considered.