Sensitivity analysis - approach for SA applied to non-linear model

Dear all,

I would like to know which approach is used in Matlab for sensitivity analysis for non-linear model with form

Y = f(X), with X a vector with n-components.

Indeed, several approaches are possible, notably variance-based methods,as Sobol total sensibility indices, or FAST method.

Other approaches to evaluate sensitivity to input parameters in model is possible, as computation of confidence intervals (eg. bootstrap monte-carlo to compute confidence interval).

Methods are more or less costly. Which are proposed to evaluate in Matlab sensitivity to input parameters, for models defined for instance in curve fitting problems? To which toolboxes are these methods bounded (curve fitting, optimization)?

Thanks and regards

[Carlos M. Velez S.]

I recommend this Matlab code for implementation of Monte Carlo method for sensitivity analysis of Simulink models: http://www.mathworks.com/matlabcentral/fileexchange/47758-sensitivity-analysis-in-simulink-models-with-monte-carlo-method