An EM algorithm for estimation in the mixture transition distribution model

The mixture transition distribution (MTD) model was introduced by Raftery to face the need for parsimony in the modeling of high-order Markov chains in discrete time. The particularity of this model comes from the fact that the effect of each lag upon the present is considered separately and additively, so that the number of parameters required is drastically reduced. However, the efficiency for the MTD parameter estimations proposed up to date still remains problematic on account of the large number of constraints on the parameters. In this article, an iterative procedure, commonly known as expectation–maximization (EM) algorithm, is developed cooperating with the principle of maximum likelihood estimation (MLE) to estimate the MTD parameters. Some applications of modeling MTD show the proposed EM algorithm is easier to be used than the algorithm developed by Berchtold. Moreover, the EM estimations of parameters for high-order MTD models led on DNA sequences outperform the corresponding fully parametrized Markov chain in terms of Bayesian information criterion. A software implementation of our algorithm is available in the library seq++at http://stat.genopole.cnrs.fr/seqpp.

• Publication year

  2008

• Venue

  Journal of Statistical Computation and Simulation

• Publication date

  2008-03-04

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1080/00949650701266666arXiv 0803.0525
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Variable Length Markov Chains: Methodology, Computing, and Software

  2004cited by this paper
• The Mixture Transition Distribution Model for High-Order Markov Chains and Non-Gaussian Time Series

  2002influential reference
• Estimation of the Mixture Transition Distribution Model

  1999influential reference
• Biological Sequence Analysis: Probabilistic Models of Proteins and Nucleic Acids

  1998cited by this paper
• Autoregressive Modelling of Markov Chains

  1995cited by this paper
• Estimation and Modelling Repeated Patterns in High Order Markov Chains with the Mixture Transition Distribution Model

  1994cited by this paper
• Some Results on the Estimation of a Higher Order Markov Chain

  1990cited by this paper
• Statistical method for predicting protein coding regions in nucleic acid sequences

  1987cited by this paper
• A model for high-order Markov chains

  1985influential reference
• ON THE CONVERGENCE PROPERTIES OF THE EM ALGORITHM

  1983cited by this paper
• Estimating the Dimension of a Model

  1978cited by this paper
• Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper

  1977cited by this paper

• hdMTD: An R Package for High-Dimensional Mixture Transition Distribution Models

  2025cites this paper
• Scenario-based model predictive control with probabilistic human predictions for human–robot coexistence

  2024cites this paper
• Leader-follower dynamics in real historical time: a Markovian test of non-linear causality between sail and steam (co-)development

  2022cites this paper
• GenMarkov: Modeling Generalized Multivariate Markov Chains in R

  2022cites this paper
• The Convex Mixture Distribution: Granger Causality for Categorical Time Series

  2021cites this paper
• Optimization of the Mixture Transition Distribution Model Using the March Package for R

  2020cites this paper
• The Profitability in the FTSE 100 Index: A New Markov Chain Approach

  2020cites this paper
• Estimation and Selection for High-Order Markov Chains with Bayesian Mixture Transition Distribution Models

  2019cites this paper
• The Profitability in the FTSE 100 Index: A New Markov Chain Approach

  2019cites this paper
• Modelling insurgent-incumbent dynamics: Vector autoregressions, multivariate Markov chains, and the nature of technological competition

  2018cites this paper
• Discovering Interactions in Multivariate Time Series

  2018influential citation
• Parsimonious higher order Markov models for rating transitions

  2018cites this paper
• Granger Causality Networks for Categorical Time Series

  2017cites this paper
• SOM clustering and modelling of Australian railway drivers’ sleep, wake, duty profiles

  2016cites this paper
• Linear models for the impact of order flow on prices. II. The Mixture Transition Distribution model

  2016cites this paper
• A New Model for Multivariate Markov Chains

  2014cites this paper
• A Markov Chain Model with High-Order Hidden Process and Mixture Transition Distribution

  2013cites this paper
• A Novel Estimation Approach for Mixture Transition Distribution Model in High-Order Markov Chains

  2009influential citation