arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1012.0996stat.MEmath.STstat.TH
收藏

马尔可夫链蒙特卡罗方法的局部相合性

Local Consistency of Markov Chain Monte Carlo Methods

Kengo Kamatani

摘要:

在本文中,我们引入了有效性(一致性)的概念,并研究了马尔可夫链蒙特卡罗方法的一些渐近性质。我们将这些结果应用于独立同分布观测的数据增广(DA)过程。更准确地说,我们表明,如果样本量和运行时间的DA程序趋于无穷大的经验分布的DA程序趋于后验分布。这是DA过程的局部属性,在某些情况下,它可能比全局属性更有助于描述其行为。使用局部性质的优点是结果的简单性和一般性。局部性质提供了有用的洞察如何构建有效的算法的问题。

英文摘要:

In this paper, we introduce the notion of efficiency (consistency) and examine some asymptotic properties of Markov chain Monte Carlo methods. We apply these results to the data augmentation (DA) procedure for independent and identically distributed observations. More precisely, we show that if both the sample size and the running time of the DA procedure tend to infinity the empirical distribution of the DA procedure tends to the posterior distribution. This is a local property of the DA procedure, which may be, in some cases, more helpful than the global properties to describe its behavior. The advantages of using the local properties are the simplicity and the generality of the results. The local properties provide useful insight into the problem of how to construct efficient algorithms.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑