Optimizing Body Weight Prediction in Bali Cattle Using Morphometric Traits and Principal Component Regression

This study aims to predict the body weight of Bali cattle using morphometric measurements through correlation analysis, multiple regression, and principal component analysis (PCA). A total of 54 female Bali cattle, consisting of 24 two-year-old and 30 three-year-old cows, were measured for body weight (BW), body length (BL), body height (BH), chest circumference (CC), chest width (CW), and chest depth (CD). Correlation analysis showed that BW had the strongest positive relationship with CC in both age groups (r = 0.87 and r = 0.91, respectively). Stepwise multiple regression produced accurate prediction models with high coefficients of determination (R2 = 0.823 for two-year-old and R2 = 0.891 for three-year-old). PCA identified CC, BL, and BH as the main contributors to body size variation, with PC1 explaining 61.97% of the total variance. Principal component regression (PCR) using PC1 and PC2 produced a robust predictive model (R2 = 0.80). These results indicate that the BW of Bali cattle can be accurately estimated using simple morphometric measurements, and PCA provides an effective approach to reduce multicollinearity and improve prediction stability in field conditions.