Skip to content
This repository has been archived by the owner on May 12, 2019. It is now read-only.

0.10.0

Past due by over 7 years 50% complete

New functionality (Gibbs sampling):

  • Gibbs factor, i.e. factor product over disjoint cliques not normalized by associated partition function,
  • construction of a variable per node in the graphical model by taking the Gibbs factor as input,
  • option for determining Jacobian correction for involved parameters,
  • tuner for adaptive Metropolis-within-Gibbs introduc…

New functionality (Gibbs sampling):

  • Gibbs factor, i.e. factor product over disjoint cliques not normalized by associated partition function,
  • construction of a variable per node in the graphical model by taking the Gibbs factor as input,
  • option for determining Jacobian correction for involved parameters,
  • tuner for adaptive Metropolis-within-Gibbs introduced by Roberts and Rosenthal,
  • adaptive Metropolis-within-Gibbs (AMWG) sampler by Roberts and Rosenthal,
  • examples of using AMWG to run simulations from OpenBUGS' "Examples, Volume 1".
Loading