Abstract
Summary: The classical approach to the analysis of data from repeated surveys, based on Patterson (1950), has recently been extended by the work of Blight and Scott (1973) and Scott and Smith (1974), by assuming that a time series relationship exists between the population parameters at different times. The purpose of this paper is to compare the efficiencies of these three approaches by computing the mean square error (MSE) of the estimators of the current mean and of the change in mean on the last two occasions.
Keywords

This publication has 5 references indexed in Scilit: