mcmc-0.8.2.0: Sample from a posterior using Markov chain Monte Carlo
Copyright2021 Dominik Schrempf
LicenseGPL-3.0-or-later
Maintainerdominik.schrempf@gmail.com
Stabilityunstable
Portabilityportable
Safe HaskellSafe-Inferred
LanguageHaskell2010

Mcmc.Proposal.Scale

Description

Creation date: Thu May 14 21:49:23 2020.

Synopsis

Documentation

scale :: Shape Double -> Scale Double -> PName -> PWeight -> Tune -> Proposal Double Source #

Multiplicative proposal.

The gamma distribution is used to sample the multiplier. Therefore, this and all derived proposals are log-additive in that they do not change the sign of the state. Further, the value zero is never proposed when having a strictly positive value.

Consider using slide to allow proposition of values having opposite sign.

scaleUnbiased :: Shape Double -> PName -> PWeight -> Tune -> Proposal Double Source #

See scale.

The scale of the gamma distribution is set to (shape)^{-1}, so that the mean of the gamma distribution is 1.0.

scaleContrarily :: Shape Double -> Scale Double -> PName -> PWeight -> Tune -> Proposal (Double, Double) Source #

See scale.

The two values are scaled contrarily so that their product stays constant. Contrary proposals are useful when parameters are confounded.