Application of gamma regression model for assessing factors affecting milk production in Holstein dairy herd

Abstract

Generalized linear models (GLMs) are an extension of the linear regression model which is a strong and flexible tool for generating relationships between predictors and response variable, GLMs are widely applied in different fields such as epidemiology, economy, finance, and veterinary medicine. Gamma regression is a type of generalized linear model that is used to model continuous response variables that are non-negative and have a skewed distribution. Reliable records of a commercial dairy farm in the Sharkia governorate of Egypt were used to collect data on 351 purebred Holstein-Friesian cows. These cows delivered between January 2018 and December 2019. The purpose of this study was the application of gamma regression model for assessing factors affecting milk production in Holstein-Friesian dairy herd by evaluating several parameters such as calving season, parity, incidence of mastitis disease, days to first insemination (DFI), days in milk (DIM) and days open (DO). The study's findings showed that the winter and autumn calving seasons, the incidence of mastitis, and the number of days to first insemination (DFI) were significant factors for the 305-day milk yield; parity, days open (DO), and days in milk (DIM) had little effect on the amount of milk produced by dairy farms.