A vanilla Rao--Blackwellization of Metropolis--Hastings algorithms

@article{Douc2011AVR,
  title={A vanilla Rao--Blackwellization of Metropolis--Hastings algorithms},
  author={R. Douc and C. Robert},
  journal={Annals of Statistics},
  year={2011},
  volume={39},
  pages={261-277}
}
  • R. Douc, C. Robert
  • Published 2011
  • Mathematics
  • Annals of Statistics
  • Casella and Robert (1996) presented a general Rao--Blackwellisation principle for accept-reject and Metropolis-Hastings schemes that leads to significant decreases in the variance of the resulting estimators, but at a high cost in computing and storage. Adopting a completely different perspective, we introduce instead a universal scheme that guarantees variance reductions in all Metropolis-Hastings~based estimators while keeping the computing cost under control. We establish a central limit… CONTINUE READING
    26 Citations

    Figures and Tables from this paper

    Using Parallel Computation to Improve Independent Metropolis–Hastings Based Estimation
    • 50
    • PDF
    Variance reduction of estimators arising from Metropolis–Hastings algorithms
    • 1
    • PDF
    Markov Chain Importance Sampling - a highly efficient estimator for MCMC
    • 5
    • PDF
    Jump Markov Chains and Rejection-Free Metropolis Algorithms
    • PDF
    Control functionals for Monte Carlo integration
    • 148
    • Highly Influenced
    • PDF

    References

    SHOWING 1-10 OF 17 REFERENCES
    Rao-Blackwellisation of sampling schemes
    • 589
    • Highly Influential
    Beyond accept-reject sampling
    • 8
    Post-Processing Accept-Reject Samples: Recycling and Rescaling
    • 30
    Limit theorems for weighted samples with applications to sequential Monte Carlo methods
    • 67
    • PDF
    Limit theorems for weighted samples with applications to sequential Monte Carlo methods
    • 113
    Simulating events of unknown probabilities via reverse time martingales
    • 30
    • PDF
    Self-regenerative Markov chain Monte Carlo with adaptation
    • 29
    • PDF
    We use asymptotic results for conditional triangular arrays of random variables given in Douc and Moulines (2008, Theorem 11)
    • 2008