非正态假定下贝叶斯动态线性模型的监控研究

摘要

Ever since, the monitoring of Bayesian Dynamic Models without normal assumptions, especially nonlinear, is a difficult problem. In this paper, by constructing the statistic based on sampling points, we realize the monitoring of Bayesian Dynamic Linear Models without normal assumptions. Furthermore, we point out that this method can be generalized to Nonlinear Bayesian Dynamic Models.%对非正态假定下贝叶斯动态模型,特别是非线性的模型的监控,一直是个难题.本文通过构建基于样本点的统计量,实现了对非正态假定下贝叶斯动态线性模型的监控.该方法也适用于非线性的贝叶斯动态模型.