Fractional Factorial Method

FractionalFactorial()

FractionalFactorial does not have any keyword arguments.

Method Details

Fractional Factorial method creates a design matrix by utilizing Hadamard Matrix and uses it to run simulations of the input model. The main effects are then evaluated by dot product between the contrast for the parameter and the vector of simulation results. The corresponding main effects and variance, i.e. square of the main effects, are returned as results for Fractional Factorial method.

API

gsa(f, method::FractionalFactorial; num_params, p_range = nothing, kwargs...)

Example

using GlobalSensitivity, Test

f = X -> X[1] + 2 * X[2] + 3 * X[3] + 4 * X[7] * X[12]
res1 = gsa(f,FractionalFactorial(),num_params = 12,samples=10)