EXPERIMENTAL-DESIGN FOR A NONLINEAR-REGRESSION MODEL IN THE ANALYSIS OF LONG-TERM SELECTION EXPERIMENTS

An experimental design problem is considered for the analysis of long- term selection experiments with nonlinear regression models. For a 3-p arametric exponential regression function whose parameters have also a reasonable biological interpretation an approximate formula for the d etermination of the necessary number of observations at each generatio ns is constructed in such a way that the half expected length of an (1 -alpha)-confidence interval for the selection limit is not greater tha n a given value. In this sense the accuracy of the parameter estimator can be described.