Sixth International Conference on Sensitivity Analysis of Model Output How to avoid a perfunctory sensitivity analysis

By Andrea Saltelli, Paola Annoni. 2010

1,035 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

An evaluation of adaptive surrogate modeling based optimization with two benchmark problems (2014, background): “, the effect of varying two parameters jointly), and total effect (including all the possible synergetic terms between that parameter and all the others) (Saltelli and Annoni, 2010).”
Scientific Opinion on good modelling practice in the context of mechanistic effect models for risk assessment of plant protection products (2014, background): “It should be noted that there are shortcomings to sensitivity analysis conducted one step at a time because they cannot assess interactions between parameters (Saltelli and Annoni, 2010) and so it may be desirable to supplement one step at a time with a global analysis (example in Figure 12, bottom ”
Characterising performance of environmental models (2013, methodology): “For more information on sensitivity analysis, see Saltelli et al. (2000), Norton (2008), Frey and Patil (2002), Saltelli and Annoni (2010), Shahsavani and Grimvall (2011), Makler-Pick et al. (2011), Yang (2011), Nossent et al. (2011), Ratto et al. (2012) and Ravalico et al. (2010).”
A Review on Global Sensitivity Analysis Methods (2014, unknown): “[74]) for estimating small indices, Jansen formula (Jansen [41], Saltelli et al.”
Forest Growth and Yield Modeling (2011, methodology): “Although sensitivity analyses are generally done “one factor at a time,” this method is highly inadequate, and other methods like regressing output variables against input variables are generally more effective (Saltelli and Annoni, 2010).”
Emulation of a complex global aerosol model to quantify sensitivity to uncertain parameters (2011, background): “However, there are two significant disadvantages of OAT tests: firstly, the fraction of parameter space sampled quickly tends to zero as the number of model inputs increases (Saltelli and Annonia, 2010).”
The effects of Labour Market Reforms upon Unemployment and Income Inequalities: an agent based model (2018, unknown): “Indeed, the SA allows to improve the identiﬁcation of the model response to changes in the parameters, thus providing clearer and more reliable propositions in policy terms (Saltelli and Annoni, 2010).”
AquaCrop: FAO's crop water productivity and yield response model (2014, methodology): “, 2013) and uncertainty analysis of different steps of the modeling process (e.g. Refsgaard et al., 2007; Saltelli and Annoni, 2010; Vanuytrecht et al., 2014a; Warmink et al., 2010).”
Why so many published sensitivity analyses are false: A systematic review of sensitivity analysis practices (2017, methodology): “We can further illustrate this with a simple example, taken from (Saltelli and Annoni, 2010).”
Image Counterfactual Sensitivity Analysis for Detecting Unintended Bias (2019, background): “• Exploring the interaction effects of attributes on the model [39, 7].”
The Future of Sensitivity Analysis: An essential discipline for systems modeling and policy support (2020, unknown): “…been commonly used much more broadly, in circumstances where it has been criticized for providing only a localized view of the problem space, especially when used in the context of investigating parameter importance in mathematical modeling (Saltelli and Annoni, 2010; Saltelli et al., 2019).”
Sobol' sensitivity analysis of a complex environmental model (2011, background): “In recent years, this broad range of possibilities made sensitivity analysis a main topic in environmental modeling literature and, particularly, in this journal (Estrada and Diaz, 2010; Ravalico et al., 2010; Saltelli and Annoni, 2010; Yang, 2011).”
Variance-based sensitivity analysis of a forest growth model (2012, methodology): “…(7) ŜTi = 1 2N ∑N j=1 (f (P)j − f (P(i)Q )j) 2 V̂(Y) (8) where ̂· · · means the estimate; f̂0 is the estimated value of the model output; P(i)Q represents all columns from P except the i-th column which is from Q, using a radial sampling scheme (Campolongo et al., 2011; Saltelli and Annoni, 2010).”