[On convergence of the iterative conditional estimation]
The iterative conditional estimation (ICE) is an iterative estimation method of the parameters in the case of incomplete data. Its use asks for relatively weak hypotheses and it can be performed in relatively complex situations, as in triplet Markov models. The aim of this Note is to express a general theorem of convergence of ICE, and to show its applicability in the problem of the estimation of the proportions in a mixture of multivariate distributions.
L'estimation conditionnelle itérative (ECI) est une méthode d'estimation itérative des paramètres dans le cas des données incomplètes. Sa mise en œuvre demande des hypothèses relativement faibles et peut être effectuée dans des situations relativement complexes, comme les modèles de Markov triplets. L'objet de cette Note est d'énoncer un théorème général de convergence de l'ECI, et de montrer son applicabilité dans le problème de l'estimation des proportions dans un mélange de lois multi-variées.
Accepted:
Published online:
Wojciech Pieczynski 1
@article{CRMATH_2008__346_7-8_457_0, author = {Wojciech Pieczynski}, title = {Sur la convergence de l'estimation conditionnelle it\'erative}, journal = {Comptes Rendus. Math\'ematique}, pages = {457--460}, publisher = {Elsevier}, volume = {346}, number = {7-8}, year = {2008}, doi = {10.1016/j.crma.2008.02.023}, language = {fr}, }
Wojciech Pieczynski. Sur la convergence de l'estimation conditionnelle itérative. Comptes Rendus. Mathématique, Volume 346 (2008) no. 7-8, pp. 457-460. doi : 10.1016/j.crma.2008.02.023. https://comptes-rendus.academie-sciences.fr/mathematique/articles/10.1016/j.crma.2008.02.023/
[1] Unsupervised statistical segmentation of non stationary images using triplet Markov fields, IEEE Trans. Pattern Analysis and Machine Intelligence, Volume 29 (2007) no. 8, pp. 1367-1378
[2] Inference in Hidden Markov Models, Series in Statistics, Springer, 2005
[3] An equivalence of the EM and ICE algorithm for exponential family, IEEE Trans. Signal Processing, Volume 45 (1997) no. 10, pp. 2613-2615
[4] Multiscale oil slick segmentation from SAR image using a vector HMC model, Pattern Recognition, Volume 40 (2007) no. 3, pp. 1135-1147
[5] Localization of shapes using statistical models and stochastic optimization, IEEE Trans. Pattern Analysis and Machine Intelligence, Volume 29 (2007) no. 9, pp. 1603-1615
[6] Unsupervised segmentation of triplet Markov chains hidden with long-memory noise, Signal Processing, Volume 88 (2008) no. 5, pp. 1134-1151
[7] The EM Algorithm and Extension, Wiley, 1997
[8] Adaptive mixture estimation and unsupervised local Bayesian image segmentation, Graphical Models and Image Processing, Volume 57 (1995) no. 5, pp. 389-399
[9] Statistical image segmentation, Machine Graphics and Vision, Volume 1 (1992) no. 1/2, pp. 261-268
[10] Random perturbations of recursive sequences with an application to an epidemic model, J. Appl. Probab., Volume 32 (1995), pp. 559-578
Cited by Sources:
Comments - Policy