Online pseudo Marginal Sequential Monte Carlo smoother for general state spaces. Application to recursive maximum likelihood estimation of stochastic differential equations

Abstract : This paper focuses on the estimation of smoothing distributions in general state space models where the transition density of the hidden Markov chain or the conditional likelihood of the observations given the latent state cannot be evaluated pointwise. The consistency and asymptotic normality of a pseudo marginal online algorithm to estimate smoothed expectations of additive functionals when these quantities are replaced by unbiased estimators are established. A recursive maximum likelihood estimation procedure is also introduced by combining this online algorithm with an estimation of the gradient of the filtering distributions, also known as the tangent filters, when the model is driven by unknown parameters. The performance of this estimator is assessed in the case of a partially observed stochastic differential equation.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [37 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02194237
Contributor : Sylvain Le Corff <>
Submitted on : Monday, August 19, 2019 - 8:48:23 PM
Last modification on : Thursday, September 19, 2019 - 5:32:50 PM

Identifiers

  • HAL Id : hal-02194237, version 1
  • ARXIV : 1908.07254

Citation

Pierre Gloaguen, Sylvain Le Corff, Jimmy Olsson. Online pseudo Marginal Sequential Monte Carlo smoother for general state spaces. Application to recursive maximum likelihood estimation of stochastic differential equations. 2019. ⟨hal-02194237⟩

Share

Metrics

Record views

15

Files downloads

23