Repeated measure analysis in the clonal evaluation in 'Conilon' coffee

Abstract

The objective of this work was to evaluate the production of fi fty clonal varieties coffee 'Conilon' (Coffea canephora), through the repeated measure methodology in order to identify and cluster them to obtain
more productive genetic materials and to model, simultaneously, the correlation between subsequent years. The infl uence of productive cycles was effi ciently regarded in the analysis by modeling the correlations between successive records of the genetic materials of fi ve years (1997 to 2001), through this statistical methodology. The repeated measure analysis showed annual oscillations over the cycles of C. canephora, related to the largest productions in the third and forth years. Moreover, the cluster analysis grouped the 50 clones into 10 groups, according to standard production oscillations identifi ed through the adopted regression model.