oakley_Fun: Oakley & O'Hagan (2004) function

In sensobol: Computation of Variance-Based Sensitivity Indices

Oakley & O'Hagan (2004) function

Description

It implements the \insertCiteOakley2004;textualsensobol function.

Usage

oakley_Fun(X)

Arguments

X	A data frame or numeric matrix where each column is a model input and each row a sample point.

Details

The function requires 15 model inputs and reads as

y=\mathbf{a}_1^T \bm{x} + \mathbf{a}_2 ^ T \sin(\mathbf{x}) + \mathbf{a}_3 ^ T \cos(\mathbf{x}) + \mathbf{x}^T \mathbf{M}\mathbf{x}\,,

where \mathbf{x}=x_1,x_2,...,x_k, k=15, and values for \mathbf{a}^T_i,i=1,2,3 and \mathbf{M} are defined by \insertCiteOakley2004;textualsensobol. The transformation of the distribution of the model inputs from U(0, 1) to N(0, 1)) is conducted internally.

Value

A numeric vector with the model output.

References

\insertAllCited

Examples

# Define settings
N <- 100; params <- paste("X", 1:15, sep = "")

# Create sample matrix
mat <- sobol_matrices(N = N, params = params)

# Compute Oakley and O'Hagan (2004) function
Y <- oakley_Fun(mat)

sensobol documentation built on May 29, 2024, 10:05 a.m.