پديدآورندگان :
Javaheri N Department of Statistics, Semnan University, Semnan, Iran , Hosseini F Department of Statistics, Semnan University, Semnan, Iran , Karimi O Department of Statistics, Semnan University, Semnan, Iran
چكيده فارسي :
Animal model is a type of generalized linear mixed-effects model, where the non-independence among data points comes from genetic and environmental effects, these form of models are named animal models. In this study we used animal model for genetic analysis number of inseminations to conception in Iranian Holstein cows and proposed the approximate Bayesian approach for the inference. Data consisted of 72, 124 records of parities 1 to 6 from 27, 113 cows having lactation between 1981 and 2007 and distributed over 15 large Holstein herds.
Approximate Bayesian estimate of direct heritability shows that about 14 percent of the variability in the number insemination resulted in conception is expressed by breeding values.
HPD confidence interval for direct heritability suggests that minimum of 12% and up to 14% of direct heritability can be expressed by the cow breeding values.
Finally we compared the results and time consuming of the Approximate Bayesian approach with the conventional Bayesian approach with MCMC algorithms. The results indicate that the computational time of the MCMC algorithm is large compared with the Approximate Bayesian approach.
