Factor and principal component analyses as alternatives to index selection.

摘要: Selections from factor and principal component analyses were compared with those the Smith-Hazel index when selecting for several switchgrass (Panicum virgatum L.) traits. The objective of this study was to examine alternatives selection. Such procedures would potentially eliminate problems selection associated indices, including errors in genetic parameter estimates difficulty assigning relative economic weights Selection performed on 1,280 plants that evaluated over 2 years at 1 location, a randomized complete block design 4 replicates. forage yield quality comparisons analysis, maximum-likelihood analysis made three sets traits (five per set) estimate repeatability comparisons. Multivariate both simple genotypic correlation matrices. Comparisons by computing Spearman's rank correlations between plant scores computed multivariate determining number selected common methods. Among methods study, had highest high matrices indicates potential using statistical method purposes. This permit breeder reduce field costs (e.g., time, labor, equipment) required obtain necessary construct indices.