Original Contribution COMPARISON OF DIFFERENT MODELS TO DESCRIBE GROWTH OF THE JAPANESE QUAIL (COTURNIX JAPONICA)

This study compared seven growth models using body weight measurements from 300 progeny obtained from unselected random bred parents. The study which lasted for 20 weeks was carried out at the University of Maiduguri Livestock Teaching and Research farm. R 2 values for the Asymptote regression, Exponential, Gompertz, Logistic, Monomolecular, Richards and Weibull models were 0.994, 0.935, 0.997, 0.997, 0.998, 0.998, 0.998 and 0.999 respectively. The Weibull model had the highest coefficient of determination (R 2 ) value (0.999) and Exponential (0.935) least. In contrast, the exponential model had the highest MSE (Mean Square Error), SD (Standard Deviation) and AIC (Akaikes Information Criterion) values while Weibull model had the least. The exponential had the poorest fit (higher MSE, SD and AIC values and lower R 2 ) while the Weibull, Gompertz and Richards models best described the data in that order (lower MSE, SD and AIC values and higher R 2 ). In order words, the nonlinear models described the data better than linear. Therefore, based on goodness of fit criteria; R 2 , MSE, SD and AIC values, the weibull model best described live weight data of the Japanese quail in Nigeria.