Bayesian analysis of twinning and ovulation rates using a multiple-trait threshold model and Gibbs sampling.

摘要: The Multiple-Trait Gibbs Sampler for Animal Models programs were extended to allow analysis of ordered categorical data using a Bayesian threshold model. algorithm is based on augmentation, where value the unobserved underlying normally distributed variable (liability) generated in each round iteration observation. several continuous and traits. Categorical traits can have any number response levels. be different trait. used analyze twinning ovulation rates from herd cattle selected rate at U.S. Meat Research Center. Data included calves born parturition lifetime cow eggs ovulated estrous cycles before first breeding as heifers. A total 6,411 calvings was recorded 2,087 cows with 83.2% single 16.8% multiple births. 19,849 ovulations 2,332 heifers 85.2% 14.8% ovulations. Mean posterior estimates heritability fraction variance accounted by permanent environmental effects (PE) .128 .103 .168 .079 rate. estimate genetic correlation .808, PE .517. Use model could more rapid improvement through improved identification selection genetically superior animals because higher scale.