Dairy cattle

Comparative Evaluation of Wood, Wilmink, Guo-Swalve, and Random Regression Models for Describing Lactation Curves in Dairy Buffalo

---

Abstract

Abstract Text: Reliable modelling of the dynamics of lactation is a necessity in genetic assessment and herd management. Four lactation curve models with four different models were made in comparison with 9,065 longitudinal lactation records of Nili-Ravi buffaloes which had 43 test-day milk yield records taken between the period of 2001 and 2024 with each model applied on the different herds in Punjab, Pakistan. The goodness-of-fit (AIC, BIC), predictive accuracy (RMSE, MSPE) and biological (peak yield, time to peak, lactation persistency) parameters were assessed as the model performance. RRM had the best AIC (2,889,524) and BIC (2,889,557), which shows the best fit, and the estimated peak yield at the range of about 45 DIM with the highest. The Wilmink model presented the best prediction error (MSPE = 13.10; RMSE = 3.59) and constant performance at lactation stages, therefore best predicting for the operations. The performance of wood and Guo-Swalve models was in the middle. The statistical significance (p < 0.01) of models difference was found to be statistically significant (p < 0.01) in terms of fit and predictive accuracy measures. These findings support the idea that the model choice needs to be purpose-specific: RRM would define the intricate biological variation best, but Wilmink is a reliable and low-bias predictor applicable when making management choices.

Keywords: 2026

How to Cite:

Nasir, K., Hyder, A. & Ahmad, S., (2026) “Comparative Evaluation of Wood, Wilmink, Guo-Swalve, and Random Regression Models for Describing Lactation Curves in Dairy Buffalo”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2295714. doi: https://doi.org/10.31274/wcgalp.24362

Rights: 1