sensitivity

sensitivity

Sensitivity analysis.

This package provides a unified framework for analyzing how uncertainty and variability in model parameters affect model outputs. It supports multiple complementary sensitivity analysis strategies, including local, sampling-based, and global methods. Together, these enable both qualitative and quantitative assessment of parameter influence on model behavior.

The available sensitivity analysis methods are implemented in the following modules:

• sensitivity.sensitivity_fast: Global sensitivity analysis using FAST (Fourier Amplitude Sensitivity Test).

• sensitivity.sensitivity_local: Local (derivative-based) sensitivity analysis around a nominal parameter set.

• sensitivity.sensitivity_morris: Global screening based on Morris elementary effects.

• sensitivity.sensitivity_sampling: Sampling-based sensitivity analysis using parameter perturbations and statistical summaries.

• sensitivity.sensitivity_sobol: Variance-based global sensitivity analysis using Sobol indices.

All methods share a common interface and data model. This allows consistent configuration, execution, and comparison of sensitivity analysis results across different techniques.