Variance based methods have assessed themselves as versatile and effective among the various available techniques for sensitivity analysis of model output. Practitioners can in principle describe the sensitivity pattern of a model Y=f ( X_1,X_2,... X_k) with k uncertain input factors via a full decomposition of the variance V(Y) of Y into terms depending on the factors and their interactions. More often practitioners are satisfied with computing just k first order effects and k total effects, the latter describing synthetically interactions among input factors. In sensitivity analysis a key concern is the computational cost of the analysis, defined in terms of number of evaluations of f ( X_1,X_2,... X_k) needed to complete the analysis, as f ( X_1,X_2,... X_k) is often in the form of a numerical model which may take long processing time. While the computational cost is relatively cheap and k-independent for estimating first order effects, it remains expensive and k-dependent for total effect indices. In the present note we compare existing and new practices for this index and offer recommendations on which to use.

SALTELLI Andrea;  ANNONI Paola;  AZZINI Ivano;  CAMPOLONGO Francesca;  RATTO Marco;  TARANTOLA Stefano; 

2010-01-01

ELSEVIER SCIENCE BV

JRC52955

0010-4655,   

www.elsevier.com/locate/cpc,    https://publications.jrc.ec.europa.eu/repository/handle/JRC52955,   

10.1016/j.cpc.2009.09.018,