Discrepancy measures for sensitivity analysis

TL;DR

This paper introduces a simple, fast, and interpretable discrepancy-based sensitivity analysis method that effectively ranks influential parameters, making sensitivity analysis more accessible to non-specialists.

Contribution

It presents a new discrepancy measure for sensitivity analysis that is easy to understand, computationally efficient, and performs comparably to traditional variance-based methods.

Findings

Discrepancy measures can effectively rank influential parameters.

The ersatz-discrepancy matches the performance of advanced algorithms.

The proposed method is faster and easier to interpret.

Abstract

While sensitivity analysis improves the transparency and reliability of mathematical models, its uptake by modelers is still scarce. This is partially explained by its technical requirements, which may be hard to understand and implement by the non-specialist. Here we propose a sensitivity analysis approach based on the concept of discrepancy that is as easy to understand as the visual inspection of input-output scatterplots. Firstly, we show that some discrepancy measures are able to rank the most influential parameters of a model almost as accurately as the variance-based total sensitivity index. We then introduce an ersatz-discrepancy whose performance as a sensitivity measure matches that of the best-performing discrepancy algorithms, is simple to implement, easier to interpret and orders of magnitude faster.