Exact IBF sampling methods in Zero-Inflated Problems

摘要：

Zero-Inflated models such as ZIP model is a common model for count data with excess zeros. When the data are upper bounded.it is usual to use ZIB model. As to Zero-Inflated problem, it is easy to split the zeros into two parts, hence becomes a missing data problem. So it is easy to get the MLE using EM algorithm, but it is hard to get the posterior density or sample from the posterior density. This article presents an exact sampling method for the Zero-Inflated model.Using the exact IBF sampling method,we can easily get samples from posterior distributions of the unknown parameters. Numerical studies show that the exact IBF sampling method is eflicient.lt can completely avoid the problem of convergence and slow convergence encountered in the iterative algorithms such as MCMC.

关键词： Zero-Inflated model EM algorithm Inverse Bayesian Formula Data Augmentation

作者: Shuqin Jia Wei Shao Quanzheng Liu

作者单位: School of Mathematics Shandong University Jinan,China School of Mechanical Engineering Shandong University Jinan,China

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Network Technology(ICCNT 2011)(2011第三届IEEE计算机与网络技术国际会议)

会议地点: 太原

会议语种:英文

页码: 463-467

在线出版日期: 2011-02-26（万方平台首次上网日期，不代表论文的发表时间）