The data set is from 234 Nelore bulls which participate of a progeny test in the period from 1996 to 2003. The expected progeny diference (EPDs) of seven economic traits, weights at 120 e 210 days of age, maternal effect (DMPP120 and DMPP210), weights and scrotal circumferences at 365 e 450 days of age, direct effects (DDP365, DDP450, DDPE365 and DDPE450), and age at first calving (DDIPP), were used in order to classify these animals in three groups and verify which EPDs showed greatest discriminating power in forming the groups. The statistical applied techniques were: k-means clusters analysis and principal components analysis. From the three groups formed, two of them stood out in relation to values of the EPDs means. Evidence of the importance of these two groups was observed in the principal component analysis that associate to them higher values of direct EPDs of weight and scrotal circumference. The two first principal components accounted for 70.22% of the total original variability. Both techniques could be an important tools to sire classification, variable discrimination, as well as to resume multivariate information, and they could be applied to help selection in animal breeding program.
breeding program; progeny test; sire classification; principal component
