Uncertainty, sensitivity and scenario analysis: how do they fit together?
Bibliographic record
Abstract
Dealing with uncertainty is becoming increasingly important in model-based decision support.Various methods have been developed in order to do this, including uncertainty, sensitivity and scenario analysis.Although these different methods serve their purpose, the availability of a large number of methods can make it difficult for practitioners to understand the similarities and differences between them and when the use of one is more suitable than another, resulting in confusion.In addition, researchers often identify with belonging to a group dealing with a particular approach, which can lead to a lack of crossfertilisation and understanding.In order to assist with bridging the gap between researchers working on different approaches to dealing with uncertainty and eliminate confusion for practitioners, the objective of this paper is to examine the relationship between uncertainty, sensitivity and scenario analysis in the context of model-based decision support, and to take the first steps towards establishing common ground between these methods and assess the contexts under which they are most suitable.This is achieved by conceptualising the various methods as different approaches to "sampling" the hyperspace of model inputs, although this is done from different perspectives and for different ends (Figure 1).It is therefore also necessary to think about the assumptions each method is making about the space being explored, and there are benefits to be gained in thinking about how best to sample the space for each purpose.The approaches identified in this conference paper provide a first level of coarse characterisations.Further refinements in categorisation is possible (with the differentiation between narrative and stress testing scenarios as a first example), and likely to be useful.There are connections to be made to other disciplines, such as philosophy and decision theory, regarding the assumptions each method makes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.074 | 0.181 |
| Meta-epidemiology (narrow) | 0.004 | 0.003 |
| Meta-epidemiology (broad) | 0.008 | 0.003 |
| Bibliometrics | 0.018 | 0.014 |
| Science and technology studies | 0.003 | 0.016 |
| Scholarly communication | 0.023 | 0.051 |
| Open science | 0.005 | 0.011 |
| Research integrity | 0.010 | 0.015 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".