Motivated by an application in the realm of climate change economics,we develop and prove the mathematical properties of a Global Sensitivity analysis (GSA) method for function-valued responses and, by exploiting the similarity between GSA and functional analysis of variance models, we employ a domain-selective nonparametric testing technique to assess the significance of the calculated sensitivity indices. The results confirm the qualitative intuitions of previous works,determining in a quantitative way the sparsity and heterogeneity of input effects and the weak impact of interactions between inputs. Moreover, we can appreciate thecomplex and non-linear time-dynamics of input effects.
Functional Data Analysis-based sensitivity analysis of Integrated Assessment Models for Climate Change Modelling
Matteo Fontana;Massimo Tavoni;Simone Vantini
In corso di stampa
Abstract
Motivated by an application in the realm of climate change economics,we develop and prove the mathematical properties of a Global Sensitivity analysis (GSA) method for function-valued responses and, by exploiting the similarity between GSA and functional analysis of variance models, we employ a domain-selective nonparametric testing technique to assess the significance of the calculated sensitivity indices. The results confirm the qualitative intuitions of previous works,determining in a quantitative way the sparsity and heterogeneity of input effects and the weak impact of interactions between inputs. Moreover, we can appreciate thecomplex and non-linear time-dynamics of input effects.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.