Sensitivity analysis and its neighbourhoods
Sensitivity analysis may assist modellers from all fields of application to improve the quality of their inference. Sensitivity Analysis is crucial both in the model construction and model interpretation phases, and is considered an important ingredient of model verification and validation.
Sensitivity analysis is the study of how the uncertainty in the output of a mathematical model or system (numerical or otherwise) can be apportioned to different sources of uncertainty in its inputs. A related practice is uncertainty analysis, which has a greater focus on uncertainty quantification and propagation of uncertainty; ideally, uncertainty and sensitivity analysis should be run in tandem.
At times, especially when the analysis feeds into decisions, a sensitivity analysis is simply not enough. Battling parties defends their stakes appealing to different models built on different premises, driven by different interests or norms. This is where sensitivity auditing comes in, to either uphold one’s assessment of to deconstruct one’s opponent one. Based on a simple seven point checklist, sensitivity auditing may be considered a “modelling of the modelling process”.