The multivariate emulator

Share:

Description

A multivariate generalization of the interpolant() function of the emulator package

Usage

1
multem(x, expt, hp,  LoF = NULL, give=FALSE, Sigmainv=NULL, ...)

Arguments

x

Points at which the function is to be estimated in the form of an object of class mdm

expt

Points at which the code has been evaluated (x_known), in the form of an object of class experiment

hp

hyperparameter object, of class mhp

give

Boolean, with TRUE meaning to return extra information and default FALSE meaning to return just the mean

Sigmainv

The inverse of the variance matrix of the observations with default NULL meaning to calculate using var.matrix()

LoF

List of regressor functions

...

Further arguments passed to var.matrix()

Details

This is the central function of the package. It is the analogue of interpolant() of the emulator package.

Author(s)

Robin K. S. Hankin

See Also

betahat_mult

Examples

1
2
3
4
5
6
data(mtoys)
d <- obs_maker(toy_mm, toy_mhp, toy_LoF, toy_beta)
ex <- experiment(toy_mm , d)

Sigmainv <- solve(var.matrix(toy_mm,hp=toy_mhp))
multem(x=toy_mm2, expt=ex, hp=toy_mhp,LoF=toy_LoF, give=TRUE)