We consider a version of the data augmentation algorithm of Tanner and Wong, which is a special case of the Gibbs sampler. Using ideas from Harris recurrence, we derive quantitative, a priori bounds on the number of iterations required to achieve convergence. Our analysis involves relating the Markov chain to an associated dynamical system.
Publié le : 1993-08-14
Classification:
Data augmentation,
Gibbs sampler,
Harris recurrence,
convergence rate,
60J10,
62F15
@article{1177005366,
author = {Rosenthal, Jeffrey S.},
title = {Rates of Convergence for Data Augmentation on Finite Sample Spaces},
journal = {Ann. Appl. Probab.},
volume = {3},
number = {4},
year = {1993},
pages = { 819-839},
language = {en},
url = {http://dml.mathdoc.fr/item/1177005366}
}
Rosenthal, Jeffrey S. Rates of Convergence for Data Augmentation on Finite Sample Spaces. Ann. Appl. Probab., Tome 3 (1993) no. 4, pp. 819-839. http://gdmltest.u-ga.fr/item/1177005366/