Ridge regression analysis and its use in animal science

In the study, the Ridge Regression (RR) method used in multiple linear regression analyze was tried out to be explained. The significance of RR method usage preferred according to the method of Least Square method (LS) in case having multicollinearity between multiple variables, was presented especially with several researches in the area of animal science. On the other hand, in the study, the multicollinearity problem, determination of multicollinearity problem, determination methods of bias stability (k) value used in RR method and application of the method and the way of interpretation of the presented values. So as to be easily explained, it was exemplified via a simple sample in the area of animal science. It was modeled by using five different body measurements of Brown Swiss cattle for the sample. In the study, in case of a multicollinearity, the determination coefficient was found low for both models obtained with LS method under the same environmental condition of the model obtained with RR analyze (R2<0,25) and the approached feature for the model which was used with a dependent and five independent variables and was not found statistically significant. Nevertheless, the usage of RR method decreased the variance inflation factor (VIF) as it was expected and decreased the standard error values belonging to coefficient of regression for independent variables sourced from multicollinearity problems to LS method. For the method choice, it was emphasized as significant that the usage of correct method in case of multicollinearity for increasing of accuracy and the necessity of attentive and selective behaviors of researchers.