gp_jeffreys function

Jeffreys prior for GP parameters (σ,ξ\sigma, \xi)

Jeffreys prior for GP parameters (σ,ξ\sigma, \xi)

For information about this and other priors see set_prior.

gp_jeffreys(pars, min_xi = -1/2, max_xi = Inf, trendsd = 0)

Arguments

  • pars: A numeric vector of length 2. GP parameters (σ,ξ\sigma, \xi).
  • min_xi: A numeric scalar. Prior lower bound on ξ\xi. Must not be -Inf because this results in an improper posterior.
  • max_xi: A numeric scalar. Prior upper bound on ξ\xi.
  • trendsd: Has no function other than to achieve compatibility with function in the evdbayes package.

Returns

The log of the prior density.