Particle learning for general mixtures
Particle learning for general mixtures
复制标题
一般混合物的粒子学习
DOI：
10.1214/10-ba525
复制
发表时间：
2010
影响因子：
4.4
通讯作者：
Matt Taddy
中科院分区：
文献类型：
--
作者：
C. Carvalho;H. Lopes;Nicholas G. Polson;Matt Taddy
本文开发了用于估计一般混合模型的粒子学习（PL）方法。该方法与其他粒子过滤方法的区别主要有两个方面。首先，每次迭代都从根据后验预测概率对粒子进行重采样开始，从而产生更有效的传播集。其次，每个粒子仅跟踪“基本状态向量”，从而减少维度推断。此外，我们描述了该方法如何应用于当前文献中感兴趣的更一般的混合模型；希望这将激励更多的研究人员采用顺序蒙特卡罗方法来拟合其复杂的基于混合的模型。最后，我们表明PL导致了用于边际似然计算和后验聚类分配的简单工具。
This paper develops particle learning (PL) methods for the estimation of general mixture models. The approach is distinguished from alternative particle ltering methods in two major ways. First, each iteration begins by resampling particles according to posterior predictive probability, leading to a more ecient set for propagation. Second, each particle tracks only the \essential state vector" thus leading to reduced dimensional inference. In addition, we describe how the approach will apply to more general mixture models of current interest in the literature; it is hoped that this will inspire a greater number of researchers to adopt sequential Monte Carlo methods for tting their sophisticated mixture based models. Finally, we show that PL leads to straightforward tools for marginal likelihood calculation and posterior cluster allocation.
影响因子：
4.4
作者：
Rodríguez A;Dunson DB
通讯作者：
Dunson DB