Application of Nonlinear Models with Random Coefficients to Growth Data

Abstract

SUMMARY The application of nonlinear random coefficient models to the analysis of growth curve data is described. The approach is further developed for the estimation of mean growth curves and their variability from mark-recapture data when the age of an animal at first capture is unknown, but the time between successive captures is known. These methods are of wide applicability as shown by the analysis of data on the growth of emus (Dromaius novaehollandiae), noisy scrub-birds (Atrichornis clamosus), and whelks (Dicathais aegaota). Simulations of a mark-recapture experiment on the rock (spiny) lobster (Panulirus argus) in Mexico, using a known growth curve, showed the effectiveness of the approach in estimating both the population's growth curve and the variability in individual growth.