The aim of this study was to estimate genetic parameters for birth and weaning weight in guinea pigs (Cavia porcellus) reared in a farm in Cieneguilla, Lima, Peru. Univariate animal model that included sex of calf, parity, season of birth, animal and litter effect was used in the genetic analyses. In addition, the effect of number of newborns was included in the model for birth weight, and the number of born alive and age at weaning were included in the model for weaning weight. Variance components were estimated by Bayesian Gibbs sampling with a chain of100 000 rounds with the first 10 000 rounds discarded for burn-in by using MCMCglmm library of R software. Heritability estimates for birth weight was 0.155 ± 0.036 and for weaning weight was 0.246 ± 0.046. These values were similar to those estimated by using REML.

The aim of this study was to estimate genetic parameters for birth and weaning weight in guinea pigs (Cavia porcellus) reared in a farm in Cieneguilla, Lima, Peru. Univariate animal model that included sex of calf, parity, season of birth, animal and litter effect was used in the genetic analyses. In addition, the effect of number of newborns was included in the model for birth weight, and the number of born alive and age at weaning were included in the model for weaning weight. Variance components were estimated by Bayesian Gibbs sampling with a chain of100 000 rounds with the first 10 000 rounds discarded for burn-in by using MCMCglmm library of R software. Heritability estimates for birth weight was 0.155 ± 0.036 and for weaning weight was 0.246 ± 0.046. These values were similar to those estimated by using REML.

Objetivo. Identificar variables relevantes del entorno de la UNALM para caracterizar diversas tendencias que Es usual que los resultados de un experimento con medidas repetidas a través del tiempo en una misma unidad experimental sea analizado utilizando el análisis de varianza de parcelas divididas (análisis clásico), pese a que en la mayoría de los casos no siempre es posible asumir independencia entre las mediciones porque los datos longitudinales normalmente presentan correlación entre las medidas hechas a una misma unidad experimental. Una alternativa de análisis, con datos provenientes de un diseño longitudinal, es utilizar un modelo lineal mixto, que incluya una estructura de correlación entre las mediciones de cada unidad experimental. En este trabajo se aplican el análisis de varianza de parcelas dividas y modelos lineales mixtos con estructura de correlación e...