Hidden Markov models with mixtures as emission distributions
From MaRDI portal
Publication:892798
DOI10.1007/S11222-013-9383-7zbMATH Open1325.62157arXiv1206.5102OpenAlexW1972250859MaRDI QIDQ892798FDOQ892798
Authors: Stevenn Volant, Caroline Bérard, Marie-Laure Martin-Magniette, Stéphane Robin 
Publication date: 12 November 2015
Published in: Statistics and Computing (Search for Journal in Brave)
Abstract: In unsupervised classification, Hidden Markov Models (HMM) are used to account for a neighborhood structure between observations. The emission distributions are often supposed to belong to some parametric family. In this paper, a semiparametric modeling where the emission distributions are a mixture of parametric distributions is proposed to get a higher flexibility. We show that the classical EM algorithm can be adapted to infer the model parameters. For the initialisation step, starting from a large number of components, a hierarchical method to combine them into the hidden states is proposed. Three likelihood-based criteria to select the components to be combined are discussed. To estimate the number of hidden states, BIC-like criteria are derived. A simulation study is carried out both to determine the best combination between the merging criteria and the model selection criteria and to evaluate the accuracy of classification. The proposed method is also illustrated using a biological dataset from the model plant Arabidopsis thaliana. A R package HMMmix is freely available on the CRAN.
Full work available at URL: https://arxiv.org/abs/1206.5102
Recommendations
- Inference in finite state space non parametric hidden Markov models and applications
- HMM with emission process resulting from a special combination of independent Markovian emissions
- Non-parametric Bayesian inference for continuous density hidden Markov mixture model
- ESTIMATING COMPONENTS IN FINITE MIXTURES AND HIDDEN MARKOV MODELS
- Hidden Markov models with state-dependent mixtures: minimal representation, model testing and applications to clustering
Classification and discrimination; cluster analysis (statistical aspects) (62H30) Markov processes: estimation; hidden Markov models (62M05) Applications of statistics to biology and medical sciences; meta analysis (62P10)
Cites Work
- MCLUST: Software for model-based cluster analysis
- Estimating the dimension of a model
- Title not available (Why is that?)
- Finite mixture models
- Consistent estimation of the order of mixture models.
- Biological Sequence Analysis
- Methods for merging Gaussian mixture components
- Large-scale multiple testing under dependence
- Selecting hidden Markov model state number with cross-validated likelihood
- Unsupervised classification for tiling arrays: chip-chip and transcriptome
Cited In (19)
- Hidden Markov models with factored Gaussian mixtures densities
- A macro-DAG structure based mixture model
- Consistent order estimation for nonparametric hidden Markov models
- The order estimation for hidden Markov models
- Hidden Markov models for longitudinal rating data with dynamic response styles
- Hidden Markov models with state-dependent mixtures: minimal representation, model testing and applications to clustering
- HMMmix
- Extended tied-mixture HMMs for both labeled and unlabeled time series data
- Modeling animal movement with directional persistence and attractive points
- A Novel Continuous Hidden Markov Model for Modeling Positive Sequential Data
- Variational Bayes approach for model aggregation in unsupervised classification with Markovian dependency
- Inference in finite state space non parametric hidden Markov models and applications
- Bayesian Spline-Based Hidden Markov Models with Applications to Actimetry Data and Sleep Analysis
- HMM with emission process resulting from a special combination of independent Markovian emissions
- Scalable Inference for Hybrid Bayesian Hidden Markov Model Using Gaussian Process Emission
- Nonasymptotic control of the MLE for misspecified nonparametric hidden Markov models
- Spline‐based nonparametric inference in general state‐switching models
- Finite mixture model of hidden Markov regression with covariate dependence
- Identifiability and consistent estimation of nonparametric translation hidden Markov models with general state space
Uses Software
This page was built for publication: Hidden Markov models with mixtures as emission distributions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q892798)
