Utilização das distribuições inflacionadas de zeros no monitoramento da qualidade do leite

Abstract

The somatic cell count (SCC) in milk provides a more programs used in the control and prevention of mastitis worldwide. Mastitis is an endemic disease which causes damage to the animal and milk quality. The objective of this work is to identify the behavior of the variable CCS, finding a probability distribution that fits the data well and observe their behavior towards the covariates age and days in milk. The use of zero inflated distributions should be variable of interest characteristic, the excess of zero, because the value zero in contamination cases is a very important observation because it can represent the best animal genetically. With the aid of the package GAMLSS from statistical software R, we can compare the fit quality of two distributions, the Zero Inflated Poisson - ZIP and Zero Inflated Negative Binomial - ZINB. We observed that the ZINB had better fit to the data.