Fourier Amplitude Sensitivity Test (FAST)

Question

I am new in the domain of sensitivity analysis, I am trying to investigate the global sensitivity analysis method FAST (Fourier Amplitude SEnsitivity testing). I read alot about this subject, starting from the original paper of this method (Study of the Sensitivity of coupled reaction systems to uncertainties in rate coefficients, 1973), till many other articles specified for the application of this method. However till now, I am not convinced how the fourier coefficients would approximate the partial variance of an input of a model f(x1,x2,...,xn)=y.

Would some one help me by proposing a referrence where could I find a simple explanation of the concept of such method.

Thank you in advance

Answer 1

I didn't work with this method, I used classical method (Monte Carlo to estimate the Sobol indices), it takes a long time, and there is an other method based on the polynomial decomposition (PCE) (reference 3),

But there are references of this work:

1-"A Quantitative Model-Independent Method for Global Sensitivity Analysis of
Model Output", A. SALTELLI, S. TARANTOLA, and K. P.-S. CHAN
2- "Importance measures in global sensitivity analysis of nonlinear models"
Toshimitsu Homma  & Andrea Saitelip
3- Metamodel-Based Sensitivity Analysis: Polynomial Chaos Expansions and
 Gaussian Processes" Loïc Le Gratiet, Stefano Marelli, and Bruno Sudret

Good luck