5
$\begingroup$

Suppose I have an absolutely continuous distribution with density $f(x)$ and I use an mcmc sampler which has accept/reject step to sample from this distribution. In the final samples, there are some singular points (i.e., samples on top of each other). Does the distribution of the samples without these singular points converge to anything meaningful?

$\endgroup$
5
$\begingroup$

Lemma 1 in Douc & Robert, 2011, Annals of Statistics, 39(1), p.262

We addressed this problem in our 2011 vanilla Rao-Blackwellisation paper. The limiting distribution of the unique simulations in the Metropolis-Hastings sequence is associated with the density $$\tilde\pi(x)\propto\pi(x)\bar{\alpha}(x)\quad\text{where}\quad\bar{\alpha}(x)=\int_{\mathcal X}\alpha(x,y)q(y|x)\,\text{d}y$$if

  • $\pi(\cdot)$ is the original target of the Metropolis-Hastings algorithm
  • $\alpha(\cdot,\cdot)$ is the Metropolis-Hastings acceptance probability
  • $q(\cdot|\cdot)$ is the Metropolis-Hastings proposal or kernel

(In the above excerpt from the paper, $\mathfrak z_i$ denotes the $i$th accepted value in the MCMC chain and $\mathfrak n_i$ the number of times it is repeated in the MCMC chain.)

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.