Gradient free parameter estimation for hidden Markov models with intractable likelihoods (Q2516386)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Gradient free parameter estimation for hidden Markov models with intractable likelihoods
scientific article

    Statements

    Gradient free parameter estimation for hidden Markov models with intractable likelihoods (English)
    0 references
    0 references
    0 references
    0 references
    31 July 2015
    0 references
    The authors focus on a maximum likelihood estimation for the static model parameters of hidden Markov models (HMMs). They do not consider the cases where the computation of the conditional likelihood density of the observation is very complex. They obtain samples from the conditonal likelihood and use an approximate Bayesian computation of the HMM.
    0 references
    approximate Bayesian computation
    0 references
    hidden Markov models
    0 references
    parameter estimation
    0 references
    sequential Monte Carlo method
    0 references
    maximum likelihood estimation
    0 references
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references