Principal component regression analysis to predict lifetime milk yield of Jaffarabadi buffaloes

Abstract

The study aims to devise most appropriate prediction model for lifetime milk production of Jaffarabadi Buffalo, based on principal components formulated on initially expressed lactation records as predictors. Lactation milk yield, lactation period and peak milk yield records of first, second and third lactations of animals under study were used of 24 years (1987 to 2010). Principal components (PCs) were derived from data set using principal component regression analysis (PCRA), the principal components were used as predictors for predicting lifetime milk yield (LTMY). Multiple linear regression models were fitted to identify the best fitted model for prediction of lifetime milk yield with the first principal component to all principal component as a predictor. The equation LTMY = 7825.8768+2.8118 (PC1) - 13.7098 (PC2) - 599.0908 (PC3) + 3.0266 (PC4) - 8.8196 (PC5) - 257.9315 (PC6) + 2.6042 (PC7) explained 98.9% variation in the estimated values with adjusted R2= 59.09% variation in the estimated values. The curve estimation analysis showing the appropriateness of first seven principal components as predictor was the most appropriate model for lifetime milk yield. These prediction equations may be helpful in selection at an early stage of Jaffarabadi Buffalo based on early part lactation records.