topsa (version 0.1.0)

topsa: Topological Sensitivity Analysis

Description

Topological Sensitivity Analysis

Usage

topsa(Ydat, Xdat, threshold.radius = rep(0.05, ncol(Xdat)), method = "Alpha")

Arguments

Ydat

numeric vector of responses in a model.

Xdat

numeric matrix or data.frame of covariables.

threshold.radius

percent of radius or sizes of triangles to keep. the homology complex. Defaults to `0.05`. complex

method

type of method to build the homology complex. Two choices are accepted: Alpha o VR (Vietoris-Rips).

Value

A list of class topsa with the following elements:

call

The function call.

Xdat

X input.

Ydat

Y output.

dimension

dimension to estimate the homology order.

threshold

cutoff level for the radius or area.

results

A list for each variable with:

threshold

threshold used to limit the radius or area.

Manifold_Area

geometrical area of the estimated manifold.

Box.Area

geometrical area of the estimated manifold.

Geometric.R2

geometric correlation between each x and y.

Geometric.Sensitivity

symmetric sensitivity index of each estimated manifold.

manifold_plot

a sf object with the estimated manifold.

Examples

# NOT RUN {
ishigami.fun <- function(X) {
A <- 7
B <- 0.1
sin(X[, 1]) + A * sin(X[, 2])^2 + B * X[, 3]^4 * sin(X[, 1])
}
X <- matrix(runif(3*50, -pi, pi), ncol = 3)
Y <- ishigami.fun(X)
estimation <- topsa(Ydat = Y, Xdat = X,method = "Alpha")
# }