The objetive of this study was to evaluate the performance of broiler genotypes using repeated measurements. Four genotypes (A, B, C, and D) and two sexes were evaluated at six ages (7, 14, 21, 28, 35, and 42 days) for average feed intake (AFI), average body weight (ABW) and feed:gain ratio (F/G) using an unbalanced incomplete blocks 4x2 factorial design. Five error co(variance) structures were tested using the MIXED procedure of SAS® for statistical analyses. Averages were estimated by least squares and compared by the Tukey-Kramer test. Quadratic profile analyses for AFI and F/G and Gompertz growth model for ABW and respective coefficients of determination were estimated by NLIN procedure of SAS®. Significant effects of some triple or double interactions were observed for all response variables. Genotypes signicantly diferred in each age and sex for AFI and ABW after 21 days of age and diferences for F/G between genotypes and sexes, in each age, were verified only at 42 days of age. Larger averages for AFI and ABW was observed for genotype D and smaller F/G was verified in genotypes C and B. Estimated profile analyses explained adequately each response variable in function of age. AFI and ABW for males of genotype D were larger after 14 days and from 28 to 42 days their performance diferred from the other genotypes. The best F/G was observed in males of the genotype C. Overall, genotypes C and B presented the best performance. The repeated measurements approach was approppriate to evaluate diferences in the performance of broiler genotypes.
genetic breeding; mixed model; MIXED procedure; NLIN procedure
