log_approx_ref_prior {RobustGaSP} | R Documentation |
The natural logarithm of the jointly robust prior (up to a normalizing constant)
Description
A function to the natural logarithm of the jointly robust prior (up to a normalizing constant).
Usage
log_approx_ref_prior(param, nugget, nugget_est, CL, a, b)
Arguments
param |
A vector of natural logarithm of inverse-range parameters and natural logarithm of the nugget-variance ratio parameter. |
nugget |
The nugget-variance ratio parameter if this parameter is fixed. |
nugget_est |
Boolean value of whether the nugget is estimated or fixed. |
CL |
Prior parameter in the jointly robust prior. |
a |
Prior parameter in the jointly robust prior. |
b |
Prior parameter in the jointly robust prior. |
Value
The numerical value of the derivative of the approximate reference prior with regard to inverse-range parameters and the nugget-variance ratio parameter. When the nugget is fixed, the derivative is on inverse-range parameters.
Author(s)
Mengyang Gu [aut, cre], Jesus Palomo [aut], James Berger [aut]
Maintainer: Mengyang Gu <mengyang@pstat.ucsb.edu>
References
M. Gu. (2016). Robust uncertainty quantification and scalable computation for computer models with massive output. Ph.D. thesis. Duke University.
M. Gu (2018), Jointly robust prior for Gaussian stochastic process in emulation, calibration and variable selection, arXiv:1804.09329.
See Also
Examples
# inputs
x<-runif(10);
n<-length(x);
# default prior parameters
a<-0.2
b<-n^{-1}*(a+1)
R0<-as.matrix(abs(outer(x,x, "-")))
CL<- mean(R0[which(R0>0)])
# compute the density of log reference prior up to a normalizing constant
param <- seq(-10,10,0.01)
prior <- rep(0,length(param))
for(i in 1:length(param)){
prior[i] <- exp(log_approx_ref_prior(param[i],nugget=0,nugget_est=FALSE,CL,a,b) )
}
# plot
plot(param,prior,type='l',
xlab='Logarithm of inverse range parameters',
ylab='Prior density up to a normalizing constant')