Pantnagar Journal of Research

Abstract

This study employed principal component analysis to examine part milk yield records from first lactation in crossbred cows, aiming to enable early selection decisions while reducing data dimensionality. Monthly milk yield measurements were collected from 529 crossbred cattle (62.5% Holstein Friesian × 37.5% Sahiwal) over three decades (1990–2019) at Govind Ballabh Pant University of Agriculture and Technology, Pantnagar. Ten lactation variables representing 30day intervals up to 300 days were analyzed using varimax rotational principal component analysis with Kaiser normalization. Two principal components emerged with eigenvalues exceeding 1, collectively explaining 77.64% of total variance. Component 1 (43.24% variance) comprised early lactation yields (days 31–120), while Component 2 (34.40% variance) represented late lactation yields (days 211–300). Varimax rotation effectively differentiated component loadings, with phenotypic correlations among traits ranging from 0.41 to 0.87, confirming strong positive associations across lactation periods. Communalities ranged from 0.52 to 0.86 across traits. The findings demonstrate that principal component analysis effectively captures lactation patterns while substantially reducing variable complexity, offering practical applications for genetic evaluation programs in crossbred cattle.