Prediction model of albumen index and height in Japanese quail eggs via external quality characteristics

This study used multiple regression analysis to estimate the relationships between egg albumen index and height and external quality traits of eggs in Japanese quails. Egg albumen index and height were selected as dependent variables, while egg weight, width, length, shape index and Haugh unit were determined as independent variables. In the multiple regression analysis, although the overall fit of the model was high, it was determined that there were multicollinearity problem among the independent variables. To solve this problem, the Principal Components Regression (PCR) method, which is widely used in the literature, was applied. With this method, the following regression equations were obtained, respectively, by using egg weight (X1, Z1), width (X2, Z2), length (X3, Z3), shape index (X4, Z4) and Haugh unit (X5, Z5) variables in estimating albumen index and height: Y ̂=(-19.95) ̂-(0.02) ̂X_1-(0.22) ̂X_2-(0.03) ̂X_3+(0.01) ̂X_4+(0.43) ̂X_5+e ̂ T ̂=(-13.21) ̂+(0.09) ̂Z_1+(0.03) ̂Z_2-(0.004) ̂Z_3-(0.005) ̂Z_4+(0.18) ̂Z_5+e ̂ The regression equations obtained were found to be statistically significant (P<0.05) and the goodness of fit of the models were determined as Detaermination coefficient (R²) = 0.86 and 0.99, respectively. Principal Components Regression (PCR) method reduced the errors caused by multicollinearity problem by decreasing the standard errors of the parameters and increased the accuracy of the model. The results show that Principal Components Regression (PCR) method effectively solved the multicollinearity problem and ensured the reliability of the model by increasing the prediction accuracy. These findings reveal that Principal Components Regression (PCR) method can be used effectively in poultry breeding and selection studies.