Use of nonlinear models to evaluate the growth curve of lambs

• Autor(a) principal: Borges, Mylena Cristina Ribeiro
• Data de Publicação: 2023
• Outros Autores: Rodrigues, Gustavo Roberto Dias, Raineri, Camila, Macedo Júnior, Gilberto de Lima, Silva, Natascha Almeida Marques da
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Caderno de Ciências Agrárias (Online)
• Texto Completo: https://periodicos.ufmg.br/index.php/ccaufmg/article/view/45002
• topic: Avaliadores de qualidade; Modelo logístico; Peso adulto; Quadrado médio do erro; Adult weight; Logistic model; Mean square error; Quality evaluators

The objective was to use non-linear regression models to evaluate the growth curve of lambs. For this, data regarding the weight and age of 70 crossbred Dorper x Santa Inês lambs born between the years 2016 to 2019 were used. The production system was intensive and semi-confined. The animal data were adjusted using non-linear Brody, Von Bertalanffy, logistic and Gompertz models. To compare the fit of the models, the adjustment quality evaluators were used: mean square error (MSE), coefficient of determination (R2) and percentage of convergence (%conv). The growth curves were made by individual adjustments. All analyzes were performed using the RStudio software, version R 4.1.2. The Logistic model was the one that best estimated the parameter a (adult weight) with 48.09 kg, while the others overestimated the biological reality of the parameter. Likewise, it presented the highest value for parameter k (maturity rate) with 0.0219. All models obtained a coefficient of determination (R²) greater than 96%. Von Bertalanffy's model had the lowest SMQ (1.61), followed by Gompetz (2.27), Logistic (2.76) and Brody (3.36). The Logistic model had the highest percentage of data convergence (87.14%), followed by Gompertz (71.43%), Von Bertalanffy (35.71%) and Brody (10%). Therefore, the logistic model showed the best fit compared to the others with adequate R², low MSE, high percentage of convergence and adequate asymptotic value, not tending to overestimate adult weight.