normal: Auxiliary functions.
In SAVE: Bayesian Emulation, Calibration and Validation of Computer Models

Auxiliary functions.

1
2
3

uniform(var.name, lower, upper)

normal(var.name, mean, sd, lower, upper)

`var.name`	The name of the variable whose prior distribution is being specified.
`lower`	The lower prior bound for `var.name`.
`upper`	The upper prior bound for `var.name`.
`mean`	The prior mean for `var.name`.
`sd`	The prior standard deviation for `var.name`.

These functions are intended as a user-friendly way of specifying the type of priors assumed for the calibration parameters in the function bayesfit.

For example, in a problem with calibration parameters named "delta1" and "shift", a uniform(0,1) prior for "delta1" and for "shift" a normal density with mean 2 and standard deviation 1 truncated to the interval (0,3), the prior should be specified as

prior=c(uniform(var.name="delta1", lower=0, upper=1),
normal(var.name="shift", mean=2, sd=1, lower=0, upper=3))

SAVE documentation built on May 2, 2019, 6:10 a.m.