Variance-based measures of sensitivity are attractive because they measure sensitivity across the whole input space (i.e. it is a global method), they can deal with nonlinear responses, and they can measure the effect of interactions in non-additive systems.
•
due to the fact that the interaction effect between e.g. Xi and Xj is counted in both STi and STj In fact, the sum of the STi will only be equal to 1 when the model is purely additive.
Green function | Graph of a function | Hash function | Schur function | Riemann zeta function | Dirac delta function | Bessel function | Monotonic function | Lateralization of brain function | function (mathematics) | Cryptographic hash function | Linear function | hash function | Gaussian function | function | Euler's totient function | Domain of unknown function | continuous function | Wave function | Quadratic function | Meijer G-function | Massieu function | Lyapunov function | Lambert W function | Gamma function | Function block diagram | Feigenbaum function | Continuous function | Computable function | Analytic function |