Stochastic versions of the EM algorithm: An experimental study in the mixture case
Stochastic versions of the EM algorithm: An experimental study in the mixture case
复制标题
DOI：
10.1080/00949659608811772
复制
发表时间：
1996-01-01
影响因子：
1.2
通讯作者：
Diebolt, J
中科院分区：
文献类型：
--
作者：
Celeux, G;Chauveau, D;Diebolt, J
我们比较了EM算法的三种不同随机版本：随机EM算法（SEM）、“模拟退火”EM算法（SAEM）和蒙特卡罗EM算法（MPEM）。我们特别关注混合分布问题。在这种情况下，我们调查这些算法的实际行为，通过密集的蒙特卡罗数值模拟和真实的数据研究。我们表明，对于一些特定的混合物的情况下，SEM算法几乎总是优于EM和“模拟退火”版本SAEM和MCEM。然而，对于一些严重重叠的混合物，这些算法都不能自信地使用。然后，SEM可以被用来作为一个有效的数据探索工具，定位显着的极大值的似然函数。在真实的数据的情况下，我们表明，SEM平稳分布提供了一个对比视图的对数似然强调明智的最大值。
We compare three different stochastic versions of the EM algorithm: The Stochastic EM algorithm (SEM), the ''Simulated Annealing'' EM algorithm (SAEM) and the Monte Carlo EM algorithm (MCEM). We focus particularly on the mixture of distributions problem. In this context, we investigate the practical behaviour of these algorithms through intensive Monte Carlo numerical simulations and a real data study. We show that, for some particular mixture situations, the SEM algorithm is almost always preferable to the EM and ''simulated annealing'' versions SAEM and MCEM. For some severely overlapping mixtures, however, none of these algorithms can be confidently used. Then, SEM can be used as an efficient data exploratory tool for locating significant maxima of the likelihood function. In the real data case, we show that the SEM stationary distribution provides a contrasted view of the loglikelihood by emphasizing sensible maxima.