The stochastic EM algorithm: estimation and asymptotic results
Feodor Nielsen, Søren
Bernoulli, Tome 6 (2000) no. 6, p. 457-489 / Harvested from Project Euclid
Text on Project Euclid
PDF
Alt PDF
The EM algorithm is a much used tool for maximum likelihood estimation in missing or incomplete data problems. However, calculating the conditional expectation required in the E-step of the algorithm may be infeasible, especially when this expectation is a large sum or a high-dimensional integral. Instead the expectation can be estimated by simulation. This is the common idea in the stochastic EM algorithm and the Monte Carlo EM algorithm.
Publié le : 2000-06-14
Classification:  EM algorithm,  incomplete observations,  simulation 
@article{1081616701,
     author = {Feodor Nielsen, S\o ren},
     title = {The stochastic EM algorithm: estimation and asymptotic results},
     journal = {Bernoulli},
     volume = {6},
     number = {6},
     year = {2000},
     pages = { 457-489},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1081616701}
}

Feodor Nielsen, Søren. The stochastic EM algorithm: estimation and asymptotic results. Bernoulli, Tome 6 (2000) no. 6, pp.  457-489. http://gdmltest.u-ga.fr/item/1081616701/