Uncertainty quantification and model validation
Uncertainty quantification and model validation characterise how uncertainty in inputs, parameters and model structure affects predictions. They establish how much confidence a model's outputs deserve. Systems Science and Modelling stewards this page for the Nexus Consortium.
- 01In briefThe short answer
- 02The methodWhat uncertainty quantification and model validation covers
- 03Reference practiceUncertainty quantification follows verification and validation practice
- 04In systemic risk workUncertainty that survives the chain
- 05Applied inPlatforms that apply this method
- 06StewardshipWho maintains this page
- 07Key termsTerms used on this page
GCRI · Institution · service level R0
The short answer.
Uncertainty quantification and model validation characterise how uncertainty in inputs, parameters and model structure affects predictions. Uncertainty quantification shows where risk estimates are robust and where they are fragile. In the Nexus Consortium, Systems Science and Modelling stewards its canonical page, and 7 other platforms apply it, including Water Nexus and Nexus Labs.
Systemic exposure in Middle East and North Africa.
Where people and networks concentrate, the exposure through which shocks travel between systems, shown for Middle East and North Africa as an example geography for this subject.
What uncertainty quantification and model validation covers.
- 01
Core methods
sensitivity analysis; Monte Carlo propagation; surrogate modelling; Bayesian calibration; model verification; multi-model ensembles.
- 02
Data and instruments
observational benchmarks; model ensembles; expert elicitation; experimental data; parameter databases.
- 03
Validation
Predictive distributions are tested for calibration and sharpness against held-out observations. Verification confirms that code solves the intended equations.
- 04
Limits
Structural uncertainty and unknown unknowns are difficult to represent. Computational cost limits exploration of high-dimensional models.
- 05
Use in risk decisions
Uncertainty quantification shows where risk estimates are robust and where they are fragile. It guides where better data or models would change decisions.
Uncertainty quantification follows verification and validation practice.
| Reference | What it sets |
|---|---|
| ASME V&V standards | Verification and validation of computational models |
| IPCC uncertainty guidance | Communication of uncertainty |
| Bayesian model calibration | Parameter inference with uncertainty |
Uncertainty that survives the chain.
When models are chained, uncertainty compounds. A hazard model's spread, an exposure dataset's error and a vulnerability curve's uncertainty must be propagated together. Without that, the final number looks more certain than any of its parts.
- Propagate uncertainty through chained models.
- Separate aleatory and epistemic uncertainty.
- Use sensitivity analysis to find dominant sources.
- Applications on Systems Modelling and Nexus Labs.
Platforms that apply this method.
Who maintains this page.
Systems Science and Modelling stewards uncertainty quantification and model validation: it keeps this description current and reviews how the method is applied across the platforms.
Terms used on this page.
Each term as this page uses it. Where a standard-setting body defines the term, the wording follows that body.
| Term | Definition |
|---|---|
| Primary uncertainty | Uncertainty about whether, where and how often events of a given size occur. |
| Secondary uncertainty | Uncertainty about the damage caused given that an event occurs. |
| Prediction interval | A range expected to contain a future observation with a stated probability. |
| Structural uncertainty | Uncertainty arising from the choice of model form rather than its parameters. |
Where this page's references lead.
The bodies and documents this page is read against. Status and editions change; each link goes to the issuing body.
Where this page leads across the Consortium.
- Uncertainty quantification for water modelsGCRIApplied in Water Nexus.
- Uncertainty quantification in laboratoriesGCRIApplied in Nexus Labs.
- Uncertainty in published assessmentsGCRIApplied in Nexus Reports.
- Uncertainty quantification for riskGCRIApplied in Systemic Risk Management.
- Uncertainty quantification and model validation for reinsurance and catastropheGRAApplied in Reinsurance and Catastrophe.
- Uncertainty quantification and model validation at AI and Financial Model RiskGRAApplied in AI and Financial Model Risk.
- Uncertainty quantificationGCRIApplied in Systems Science and Modelling.
- Uncertainty quantification and model validation for actuarial and quantitative riskGRAApplied in Actuarial and Quantitative Risk.
Questions and answers.
Why does one institution host every method page?
How do I suggest a correction to a method page?
Around GCRI.
Start with an enquiry.
The current service level is R0: information and enquiries. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.