Create an object representing inverse chi-squared priors with possibly modelled degrees of freedom and scale parameters
pr_invchisq(df = 1, scale = 1)An environment representing the specified prior, for internal use.
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 modelled scale parameter of the inverse chi-squared
distribution is (a scalar parameter) common to all n parameters.