f.prime: First derivative of the posterior of a spatial GEV model with
respect to a random effect in the location parameter.
Description
This internal function gives the first derivative of a spatial GEV
model with respect to a random effect parameter in order to made a
focused update of that variate.
Usage
f.prime(tau, tau.hat, varsigma, xi, kappa, R)
Arguments
tau
Current value of the random effect
tau.hat
Conditional mean of that random effect given the Gaussian process
and the other random effects
varsigma
Conditional variance of the random effect given the Gaussian process
xi
Current shape parameter for that site
kappa
Current precision parameter for that site
R
Current vector of differences between the observations and the
linear part of the location parameter at the site.
Value
A scalar giving the first derivative