GROWTH CHARACTERISTICS OF NOILER CHICKENS AS DETERMINED BY NONLINEAR ALGORITHMS

Abstract

This study sought to determine the adequacy of two mathematical functions in modelling growth characteristics of male and female Noiler chickens. The study was conducted at the Livestock Teaching and Research Farm, University of Maiduguri. Body weights were obtained weekly for 20 weeks from two hundred (200) Noiler chickens and fitted to the Gompertz and Logistic models. The results of the study showed significant difference (P<0.05) between males and females only at 20 weeks of age, with values of 3002.7 and 2718.7, respectively. The asymptomatic weight of the models revealed that Gompertz model had higher values for both male and female Noilers than the logistic. The coefficient of determination (R²) of Gompertz model (0.8122) was higher than that of logistic (0.8103) for male Noilers. Similar trend occurred in female Noilers with Gompertz having a higher R² . The Akaike information criteria (AIC) value of 6927.5 for Gompertz model was lower compared that of logistics (6933.0) in the male Noilers. Gompertz also had lower value of AIC for female Noilers compared to the Logistic. The study reveal that the Gompertz model was the better growth model for explaining the growth patterns of both male and female Noilers.