Application of partial least squares regression to predict feed efficiency based on feeding behaviour patterns in confined beef steers fed a concentrate diet

摘要

Previous research has shown that confined beef steers with feed-efficient (low RFI) phenotypes have lower (P < 0.01) frequency and duration of bunk visit (BV) and meal events (Lancaster et al., 2009), and reduced (P < 0.05) day-to-day variation of feeding behaviour patterns (Parsons et al., 2018) than high-RFI steers. Partial least squares regression is a statistical method designed to handle multivariate datasets comprised of variables with substantial collinearity. The objective of this study was to calibrate and validate PLSR-based models to predict feed efficiency of finishing beef steers using feeding behaviour traits as independent variables.