The Alive Particle Filter and Its Use in Particle Markov Chain Monte Carlo
The Alive Particle Filter and Its Use in Particle Markov Chain Monte Carlo
复制标题
活性粒子滤波器及其在粒子马尔可夫链蒙特卡罗中的应用
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
Xiaole Zhang
中科院分区：
文献类型：
--
作者：
P. Moral;A. Jasra;Anthony Lee;C. Yau;Xiaole Zhang
在下面的文章中，我们研究了一种粒子滤波器来近似具有指示电位的Feynman-Kac模型，并在马尔可夫链蒙特卡罗（MCMC）中使用该算法来学习模型的静态参数。这样的模型的例子包括近似贝叶斯计算（ABC）后验与隐马尔可夫模型（HALTH）或罕见事件的问题。这种模型需要使用先进的粒子滤波器或MCMC算法来执行估计。现有粒子滤波器的缺点之一是它们可能“崩溃”，因为算法可能由于指示器电位而提前终止。在这篇文章中，使用新开发的局部自适应粒子滤波器的特殊情况下，我们使用的算法，可以处理后一个问题，同时引入一个随机成本每时间步。特别是，我们展示了如何使用这种算法可以在MCMC，使用粒子MCMC。它是建立，当不考虑计算时间，当新的MCMC算法被应用到一个简化的模型，它有一个较低的渐近方差相比，一个标准的粒子MCMC算法。数值算例的ABC近似的障碍。
In the following article, we investigate a particle filter for approximating Feynman–Kac models with indicator potentials and we use this algorithm within Markov chain Monte Carlo (MCMC) to learn static parameters of the model. Examples of such models include approximate Bayesian computation (ABC) posteriors associated with hidden Markov models (HMMs) or rare-event problems. Such models require the use of advanced particle filter or MCMC algorithms to perform estimation. One of the drawbacks of existing particle filters is that they may “collapse,” in that the algorithm may terminate early, due to the indicator potentials. In this article, using a newly developed special case of the locally adaptive particle filter, we use an algorithm that can deal with this latter problem, while introducing a random cost per-time step. In particular, we show how this algorithm can be used within MCMC, using particle MCMC. It is established that, when not taking into account computational time, when the new MCMC algorithm is applied to a simplified model it has a lower asymptotic variance in comparison to a standard particle MCMC algorithm. Numerical examples are presented for ABC approximations of HMMs.
DOI：
10.1214/14-aap1022
发表时间：
2015
期刊：
The Annals of Applied Probability
影响因子：
--
作者：
Andrieu C
通讯作者：
Andrieu C
影响因子：
1.8
作者：
C. Andrieu;M. Vihola
通讯作者：
C. Andrieu;M. Vihola