A game theoretic approach to design mating programs for livestock

Abstract

In this study, we proposed a new game theoretic method to design a mating program. The index and game theoretic methods were applied with calculated breeding values using pedigree on two different data sets whose economic traits consist of negative (milk yield and fat percentage) and positive (birth and weaning weight) genetic correlated data. For the negative genetic correlated data set, even if total expected benefits were equal for two methods, mating programs were changed, and the coefficient of variation obtained using the game theoretic method was smaller than that of the index method. This result showed that the expected breeding value will be more homogeneous in the next generation if the game theoretical approach is used. For the positive genetic correlated data set, the total expected benefit obtained from the index selection was a bit higher than the expected benefit obtained from the game theory. In terms of the coefficient of variation, selection of the index method provides 25% more homogeneous next generation flock structure than the game theoretic approach. When the results examined, it is clear that more studies should be done using game theoretical modeling, which is a new approach for animal mating design.