Create an object containing information about inverse chi-squared priors with possibly modeled degrees of freedom and scale parameters
pr_invchisq(df = 1, scale = 1, n = NULL, post = FALSE)degrees of freedom parameter. This can be a numeric scalar or
vector of length n, the dimension of the parameter vector.
Alternatively, for a scalar degrees of freedom parameter,
df="modeled" or df="modelled" assign a default (gamma) prior
to the degrees of freedom parameter. For more control of this gamma prior a
list can be passed with some of the following components:
shape parameter of the gamma distribution
rate parameter of the gamma distribution
"RW" for random walk Metropolis-Hastings or "mala" for Metropolis-adjusted Langevin
(starting) scale of Metropolis-Hastings update
whether to adapt the scale of the proposal distribution during burnin to achieve better acceptance rates.
scalar or vector scale parameter. Alternatively,
scale="modeled" or scale="modelled" puts a default
chi-squared prior on the scale parameter. For more control on this
chi-squared prior a list can be passed with some of the following components:
degrees of freedom (scalar or vector)
scale (scalar or vector)
whether the modeled scale parameter of the inverse chi-squared
distribution is (a scalar parameter) common to all n parameters.
dimension, if known. For internal use only.
whether conditional posterior sampling function should be created. For internal use only.
An environment with information about the prior and possibly conditional posterior distribution(s), to be used by other package functions.