Using factor analysis in beef cattle breeding

Studies of the structure of the interaction of signs with the help of factor analysis is based on the idea of the complex nature of the phenomenon being studied, expressed in the interdependence of individual signs, determined by “internal” hidden causes, forming the specifics of a property, quality. We used the factor analysis method to study the producers’ assessments of the quality of the offspring and to test their bullsons for their own productivity of the Kalmyk and Hereford cattle breeds. The analysis included material from the work of pedigree reproducers and plants in the Kalmyk and Hereford breeds of the Rostov Region. The total number of animals included in the treatment was 96 manufacturing bulls and 272 sons bulls. As a result of the studies, the proposed breeding indices on one basis are quite suitable for the preliminary selection of bullcalves in commodity farms or in pedigree farmers for breeding beef cattle. In order to get a final assessment, we compiled new breeding indices, including two (live weight at 8 and 15 months, live weight at 15 months and average daily gain from 8 to 15 months), as well as three characteristics (live weight in 8, 15 months and average daily gain). These signs are objective, easy to account. According to the estimates obtained for individual bull-sons, it is possible by summing up the values of new factors to identify the bulls’ assessment of the quality of the offspring. Analysis of assessment materials is characterized by high reliability (P > 0.999).


Introduction
Studies of the structure of the interaction of signs with the help of factor analysis is based on the idea of the complex nature of the phenomenon being studied, expressed in the interdependence of individual signs, determined by "internal" hidden causes, forming the specifics of a property, quality. Factor analysis is used to concentrate the source information, including a large number of signs of the analyzed compression phenomenon using common factors, expressing in less number of more capacious internal characteristics, i.e. allows you to reduce a large amount of data to possibly whiter than simple concentration with minimal loss of information. The most capacious characteristics obtained, called factors, are not directly measurable. They are behind the scenes of the phenomenon under study, serve as its background and can only be determined as a result of analysis.
The task of factor analysis is to find a simple structure that would accurately reflect and reproduce real, existing in nature dependencies. Moreover, the form and amount of experimental data strongly influence the implementation of the principle of simple structure. A one-sided approach to the selection of signs inevitably leads to a mismatch of the phenomenon being studied. The result of the analysis is determined by the formulation of the entire study.
The goal was to construct a new factor that summarizes the initial information of all the characters from all the signs of testing the bulls for their own productivity, on which the bulls' assessment of the quality of the offspring is based. At the same time, we proceeded from the fact that these signs to some extent correlate with each other. This means that either they mutually define each other, or the relationship between them is determined by some specific quantity that cannot be directly measured. This proposal is in many cases real. The factor analysis model just determines this. Using factor analysis, we determined a value that, like a sponge, absorbed the values of all the signs of an assessment. For analysis, we took the first factor (the main component), which covers the largest set of features, their largest scatter (variance). At the same time, the coverage level of the variance of the signs of the assessment was set at least 95% [1][2][3][4][5][6][7][8][9][10][11][12].

Materials and methods
For research by factor analysis, we took materials on the evaluation of producers on the quality of offspring and tests of their bull sons on their own productivity of the Kalmyk and Hereford breeds. The analysis included material from the work of pedigree reproducers and plants in the Kalmyk and Hereford breeds of the Rostov Region. The study used data from the appraisal of cattle herds of Kalmyk and Hereford breeds. Consolidated valuation statements were submitted to the tribal department of the Ministry of Agriculture and Food of the Rostov Region.

Conclusion
The proposed breeding indices on one basis are quite suitable for the preliminary selection of calves in commodity farms or in pedigree breeding for beef cattle. In order to get a final assessment, we compiled new breeding indices, including two (live weight at 8 and 15 months., Live weight at 15 months. And average daily growth in the period from 8 to 15 months.), As well as three characteristics (live weight in 8, 15 months and average daily gain). These signs are objective, easy to account. According to the estimates obtained for individual bull-sons, it is possible by summing up the values of new factors to identify the bulls' assessment of the quality of the offspring. Analysis of the assessment materials is characterized by high reliability (P> 0.999).
Thus, using factor analysis and the multiple linear regression method, the breeder's subjectivity, difficulties in accounting for feed intake, etc. are completely eliminated. The whole work is simplified. The assessment factors obtained by two or three characteristics most fully characterize the breeding value of bulls and bulls, compared to assessment on one basis, and they should probably be used in those farms where in-depth pedigree work (breeding plants) is conducted. Such breeding indices can be developed by assessing the heifers' own productivity and, on this basis, select producers who give high-quality sons and daughters. In this case, the selection will be based on theoretical, sound assumptions, and not on the intuition of breeders, which allows you to turn selection into a technological process with its strict regulation. The selection process will become collective creativity. Technologicalization of it will allow equipping specialists with a reliable means of increasing the productive and fiery qualities of beef cattle. Breeding will turn from art into a breeder's tool.