Decision Analysis On Selection Index

Selection index method has been widely applied in breeding since it was advanced. However, the advantages of it were considerably reduced when there were too much individuals to select. If every trait's general effect on aggregate genotype can be analyzed, and primary selection trait together with auxiliary, restricted and primary restricted traits can be analyzed, breeders can select not from value of selection index but from individuals'phenotypes.This paper makes an attempt towards this aim by showing the decision analysis on selection index. The path analysis of selection index will be completed, and then the decision analysis of selection index will be completed by applying decision method of path analysis. Show a way to judge primary selection trait together with auxiliary, restricted and primary restricted traits. Thus selection can be completed by selecting individuals'phenotypes direct but not by calculating every individual's value of selection index.We demonstrated that comprehensive selection index can be transformed to linear multi-regression analysis between traits and aggregate genotype. Further more, we demonstrated that comprehensive selection index can be transformed to path analysis and completed the path analysis of it. Restricted selection index was treated as comprehensive selection index with changed electronic weight. Path analysis of restricted selection index was also completed. We found the essence of path analysis of selection index is analysis of correlative heritability between traits and aggregate genotype.Decision analysis on comprehensive selection index was accomplished. Decision coefficient of every trait was defined first, which represents its general effect on aggregate genotype, including its direct effect on aggregate genotype, its indirect effect on aggregate genotype through other correlative traits and other correlative traits'indirect effect on aggregate genotype through it. Calculating decision coefficient of every trait and arranging all of them. Traits with positive decision coefficients and negative decision coefficients can be considered as auxiliary traits and restricted traits separately. Especially trait with maximal positive decision coefficient can be considered as primary selection trait and trait with minimal negative decision coefficient be considered as primary restricted trait. Thus, primary selection trait, auxiliary traits, restricted traits and primary restricted trait be ascertained by arranging decision coefficients. Selection can be completed by selecting individuals'...