含不可忽略缺失数据非线性再生散度模型参数的Bayes估计

Bayes Estimation of Nonlinear Reproductive Dispersion Models with Nonignorable Missing Mechanism

给出协变量带有不可忽略缺失数据的非线性再生散度模型的Bayes方法,缺失数据机制由Logistic回归模型来确定.Gibbs抽样技术和Metropolis-Hastings算法(简称MH算法)用来得到模型参数、缺失数据机制中回归系数的联合Bayes估计,并用实例加以说明.

Bayesian method is developed to analyze nonlinear reproductive dispersion mod-els in which the covariate variables may be missing with nonignorable missingness mechanism. The missingness is specified by a logistic regression model.the Gibbs sampler and the MH algorithm is used to obtain the joint Bayesian estimates of parameters.A real example are used to illustrate the methodology.