On the adequacy of variational lower bound functions for likelihood-based inference in Markovian models with missing values

14Citations
Citations of this article
10Readers
Mendeley users who have this article in their library.

Abstract

Variational methods have been proposed for obtaining deterministic lower bounds for log-likelihoods within missing data problems, but with little formal justification or investigation of the worth of the lower bound surfaces as tools for inference. We provide, within a general Markovian context, sufficient conditions under which estimators from the variational approximations are asymptotically equivalent to maximum likelihood estimators, and we show empirically, for the simple example of a first-order autoregressive model with missing values, that the lower bound surface can be very similar in shape to the true log-likelihood in non-asymptotic situations.

References Powered by Scopus

The Mean Field Theory in EM Procedures for Markov Random Fields

268Citations
N/AReaders
Get full text

The Mean Field Theory in EM Procedures for Blind Markov Random Field Image Restoration

105Citations
N/AReaders
Get full text

Variational learning in nonlinear gaussian belief networks

58Citations
N/AReaders
Get full text

Cited by Powered by Scopus

Bayesian methods for neural networks and related models

127Citations
N/AReaders
Get full text

Convergence properties of a general algorithm for calculating variational Bayesian estimates for a normal mixture model

87Citations
N/AReaders
Get full text

Unsupervised hybrid feature extraction selection for high-dimensional non-Gaussian data clustering with variational inference

48Citations
N/AReaders
Get full text

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Cite

CITATION STYLE

APA

Hall, P., Humphreys, K., & Titterington, D. M. (2002). On the adequacy of variational lower bound functions for likelihood-based inference in Markovian models with missing values. Journal of the Royal Statistical Society. Series B: Statistical Methodology, 64(3), 549–564. https://doi.org/10.1111/1467-9868.00350

Readers over time

‘11‘12‘13‘14‘16‘17‘19‘2100.511.52

Readers' Seniority

Tooltip

Professor / Associate Prof. 3

33%

PhD / Post grad / Masters / Doc 3

33%

Researcher 2

22%

Lecturer / Post doc 1

11%

Readers' Discipline

Tooltip

Earth and Planetary Sciences 2

29%

Agricultural and Biological Sciences 2

29%

Mathematics 2

29%

Psychology 1

14%

Save time finding and organizing research with Mendeley

Sign up for free
0