Inference for High-Dimensional Mixture Transition Distribution Models
Coerce an EM fit to an MTD model
Counts sequences of length d+1 in a sample
The total variation distance between distributions
Estimated transition probabilities
A tibble containing sample sequence frequencies and estimated probabil...
The Bayesian Information Criterion (BIC) method for inference in MTD m...
The CUT method for inference in MTD models
The Forward Stepwise (FS) method for inference in MTD models
Forward Stepwise and Cut method for inference in MTD models
Methods for objects of class "hdMTD"
Inference in MTD models
Accessors for objects of classes "MTD", "MTDest", and "hdMTD"
Methods for objects of class "MTD"
Methods for objects of class "MTDest"
EM estimation of MTD parameters
Creates a Mixture Transition Distribution (MTD) Model
Oscillations of an MTD Markov chain
Perfectly samples an MTD Markov chain
Plot method for MTD objects
Plot method for MTDest objects
Predictive probabilities for MTD / MTDest
Estimates parameters in Mixture Transition Distribution (MTD) models, a class of high-order Markov chains. The set of relevant pasts (lags) is selected using either the Bayesian Information Criterion or the Forward Stepwise and Cut algorithms. Other model parameters (e.g. transition probabilities and oscillations) can be estimated via maximum likelihood estimation or the Expectation-Maximization algorithm. Additionally, 'hdMTD' includes a perfect sampling algorithm that generates samples of an MTD model from its invariant distribution. For theory, see Ost & Takahashi (2023) <http://jmlr.org/papers/v24/22-0266.html>.