Prediction Model of Albumen Index in Duck Eggs via External Egg Quality Characteristics in Case of Multicollinearity

This study used multiple regression analysis to estimate the relationships between duck egg albumen index and external quality traits of eggs. Egg albumen index was selected as the dependent variable, and egg weight, width, length, shape index and Haugh unit were determined as independent variables. In the multiple regression analysis, it was determined that the overall fit of the model was quite high, but there were multicollinearity problem among the independent variables. In order to solve this problem, Ridge regression method, which is widely used in the literature, was applied. In order to determine the albumen index, egg weight (X1), width (X1), length (X3), shape index (X4) and Haugh unit (X5) variables were used, R ̂=(+18.2029) ̂+(0.1362) ̂X_1-(0.5736) ̂X_2-(0.1596) ̂X_3+(0.0262) ̂X_4+(0.1921) ̂X_5+e ̂. The regression equation was obtained and found to be statistically significant (P