Predicting Body Weight from Body Measurements of Palla Strain of Nellore Sheep Using Ridge and Stepwise Regression Models
Killana Eswar Bhavani *
Department of Animal Genetics and Breeding, College of Veterinary Science, Proddatur-516360, Sri Venkateswara Veterinary University, Andhra Pradesh, India.
Sunkara Vani
Department of Animal Genetics and Breeding, College of Veterinary Science, Proddatur-516360, Sri Venkateswara Veterinary University, Andhra Pradesh, India.
Kopparthi Sakunthala Devi
Department of Animal Genetics and Breeding, College of Veterinary Science, Proddatur-516360, Sri Venkateswara Veterinary University, Andhra Pradesh, India.
Punuru Pandu Ranga Reddy
Department of Animal Genetics and Breeding, College of Veterinary Science, Proddatur-516360, Sri Venkateswara Veterinary University, Andhra Pradesh, India.
Bhumireddy Jaya Madhuri
Department of Animal Genetics and Breeding, School of Veterinary and Animal Sciences, Centurion University of Technology and Management, R. Sitapur, Paralakhemundi, Gajapati, Odisha-761211, India.
Shaik Mohammad Siraj
Department of Animal Genetics and Breeding, College of Veterinary Science, Proddatur-516360, Sri Venkateswara Veterinary University, Andhra Pradesh, India.
*Author to whom correspondence should be addressed.
Abstract
The main objective of the research was to estimate body weight of Palla sheep from body measurements using ridge (RR) and stepwise regression (SR) models. Data pertaining to body weight (BW) and eight linear body measurements (BM) specifically body length (BL), height at withers (HW), chest girth (CG), paunch girth (PG), tail length (TL), face length (FL), face width (FW), and ear length (EL) were recorded from a total of 560 Palla sheep in the Nellore district of Andhra Pradesh. The optimal model for predicting body weight was selected based on the variance inflation factor (VIF), mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R²). The results indicated that both stepwise regression (SR) and ridge regression (RR) models provided a good fit to the data; however, the SR model demonstrated a marginal improvement over the RR model with a R2 value of 0.932 in predicting Palla sheep body weight from BL, FL, HW and CG measurements. These findings highlight the effectiveness of step-wise regression model in using simple, non-invasive morphometric measurements to estimate live body weight, which can aid in selection, management, and breeding decisions under field and farm conditions.
Keywords: Stepwise regression, ridge regression, body biometrics, body weight, palla sheep