This study aimed to evaluate the minimum number and interval of weightings to estimate the asymptotic weight (A) and maturing rate (k) parameters of the growth curve of cattle. Parameters A and k of the Richards non-linear model and two data sets of weights of cattle from 1 to 3,601 days were simulated. The first data set consisted of 13 to 121 weights obtained every 30 days and the second one consisted of weightings at intervals of 30 to 150 days. The decrease in number of weightings used to estimate A and k reduced the correlations between estimated and simulated parameters and increased the bias and the residuals mean square. The increase in the interval of weightings did not affect the estimates of A, but had a damaging effect on the estimates of k. Weightings every 30 days and over four years, every 60 days and over five years or every 90 days and over six years, at least, are sufficient to estimate the parameters A and k of the Richards non-linear model for cattle.
asymptotic weight; maturing rate; non-linear models; Richards
