Estimating multivariate latent-structure models
Abstract
A constructive proof of identification of multilinear decompositions of multiway arrays is presented. It can be applied to show identification in a variety of multivariate latent structures. Examples are finite-mixture models and hidden Markov models. The key step to show identification is the joint diagonalization of a set of matrices in the same nonorthogonal basis. An estimator of the latent-structure model may then be based on a sample version of this joint-diagonalization problem. Algorithms are available for computation and we derive distribution theory. We further develop asymptotic theory for orthogonal-series estimators of component densities in mixture models and emission densities in hidden Markov models.
Cite
@article{arxiv.1603.09141,
title = {Estimating multivariate latent-structure models},
author = {Stéphane Bonhomme and Koen Jochmans and Jean-Marc Robin},
journal= {arXiv preprint arXiv:1603.09141},
year = {2016}
}
Comments
Published at http://dx.doi.org/10.1214/15-AOS1376 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)