generate.toy.observations {approximator}

Er, generate toy observations

Description

Generates toy observations on four levels using either internal (unknown) parameters and hyperparameters, or user-supplied versions.

Usage

generate.toy.observations(D1, subsets, basis.fun, hpa = NULL, betas = NULL,
export.truth = FALSE)

Arguments

D1

Design matrix for level 1 code

subsets

Subset object

basis.fun

Basis function

hpa

Hyperparameter object. If NULL, use the internal (true but unknown) hyperparameter object

betas

Regression coefficients. If NULL, use the internal (true but unknown) regression coefficients

export.truth

Boolean, with default FALSE meaning to return synthetic observations and TRUE meaning to return the actual hyperparameters and coefficients.

Author(s)

Robin K. S. Hankin

References

M. C. Kennedy and A. O'Hagan 2000. “Predicting the output from a complex computer code when fast approximations are available” Biometrika, 87(1): pp1-13

Examples

data(toyapps)
generate.toy.observations(D1=D1.toy, subsets=subsets.toy, basis.fun=basis.toy)