I'm a bit confused with the MCMC sampling. It is used to approximate an unknown posterior distribution. Suppose 1000 proposed samples are accepted. How to use these samples next? How to approximate the posterior distribution?

  • 2
    $\begingroup$ In drawing samples from the posterior you already have approximated it. You'll then want to use these samples to visualise the distribution, typically with a histogram plot or a kde plot of the samples. $\endgroup$
    – Ralph
    Apr 5, 2019 at 14:35
  • $\begingroup$ also note that if a proposal was not accepted you want to keep the previous sample. i.e. just saving the accepted proposals won't leave you with samples from the target distribution $\endgroup$
    – Sam Mason
    Apr 9, 2019 at 15:20
  • $\begingroup$ bayesplot might give some examples of things you can do with the samples. some of the plots are a bit specific to their their MC sampler, but a lot of thought has been put into the package $\endgroup$
    – Sam Mason
    Apr 9, 2019 at 15:27


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.