Homogeneidade e heterogeneidade de variância residual em modelos de regressão aleatória sobre o crescimento de caprinos Anglo-Nubianos

Abstract

Resumo -O objetivo deste trabalho foi comparar modelos de regressão aleatória com diferentes estruturas de variâncias residuais, na estimação dos componentes de covariância e parâmetros genéticos de características de crescimento de caprinos. A função de regressão por meio de polinômios ortogonais de Legendre de ordem quatro foi usada para modelar a trajetória média de crescimento animal, e de ordem três, para modelar os efeitos genéticos aditivos direto e materno e de ambiente permanente. Diferentes estruturas de variâncias residuais foram consideradas, por meio de classes de uma a sete variâncias residuais, e funções de variâncias com uso dos polinômios ordinários e de Legendre, com ordens que variaram de linear até a do quarto grau. Os modelos foram comparados pelo teste de razão de verossimilhança, o critério de informação de Akaike e o critério bayesiano de Schwarz. Os modelos com funções de variâncias residuais foram superiores aos de classes de variância. O polinômio ordinário de terceira ordem apresentou melhores resultados. Os parâmetros genéticos são infl uenciados pelo ajuste da variância residual, no entanto, os estimados pelos modelos que consideraram classes de variância, polinômio ordinário e polinômio de Legendre, de terceira ordem, são semelhantes. Termos para indexação: função de covariância, parâmetro genético, polinômio de Legendre, polinômio ordinário. Homogeneity and heterogeneity of residual variance in random regression models on growth trajectory of Nubian goatsAbstract -The objective of this work was to compare random regression models with different structures of residual variances, in the estimate of covariance components and genetic parameters for growth traits in goats. Regression functions using Legendre orthogonal polynomials of the fourth order were used for modeling animal growth trajectory, and polynomial of third order for modeling direct and maternal additive genetic effects, and permanent environmental effect. Different residual variance structures were considered using alternatively step functions (from one to seven different classes of residual variances) or variance functions using ordinary and Legendre polynomials, with orders of fi t from the fi rst to fourth order. Models were compared by likelihood ratio test, Akaike's information criterion and Schwarz's bayesian criterion. Models including residual variance functions showed better results than those including classes of variance. The ordinary polynomial of third order showed better results than the other models. Genetic parameters are affected by different residual variance structures in the models, however, genetic parameters estimates using four classes of residual variance, ordinary polynomial and Legendre polynomial of third order are similar. Index terms: covariance function, genetic parameter, Legendre polynomial, ordinary polynomial. IntroduçãoA utilização de metodologias que possam estimar os parâmetros genéticos dos animais, de forma mais acurada, pode contribuir para o aumento dos ganhos genéticos obtidos c...

Evaluation of alternative schemes for recording body weights in meat-type quails by using random regression

“…These results can be attributed to the lower number of records at the final approximation of the age range, which might hamper the partition of total variance in the correct proportions for each random effect (Sousa et al, 2008).…”

Selection Indices and Multivariate Analysis Show Similar Results in the Evaluation of Growth and Carcass Traits in Beef Cattle

“…The present status of cattle breeding and selection in this country matches the methodological summary given above. There have been several efforts to estimate covariance and genetic parameters and to predict breeding values in Nellore cattle [ 3 , 7 , 21 , 22 ] with no reports on any multivariate approach to selection. Over the past 30 years, Brazilian cattle breeders have widely adopted the use of animal breeding values rather than phenotypes in the selection process.…”

Random regression models in the evaluation of the growth curve of Simbrasil beef cattle

“…Marques et al (1999), working with Simmental beef cattle, have verified increases in residual variances as animals grew old. This increase may be due to a scale effect in the weight of the animals, which could cause bias in the evaluations if a model of homogeneity of residual variance is adopted (Sarmento et al, 2010), i.e., the use of such model means that variances attributed to random effects are not partitioned correctly (Sousa et al, 2008). The objective of the present study was to estimate the types and orders of random effects of (co)variance functions at the description of the growth trajectory of Simbrasil beef cattle using RRM adjusted by Legendre polynomials.…”