Declared linear sensitivity exercise

Method, inputs and limits

Evaluates a supplied, dimension-checked linear expression at low, base and high input settings; calculates its exact independent-box range. Method version 0.2.0. Infer probabilities, validate coefficients, run arbitrary code or treat a hypothetical expression as a causal model.

Visible mathematics / decision rules

output = intercept + Σ coefficient_i × input_i

low/base/high settings change all declared variables together

box minimum/maximum select each endpoint according to the sign of its coefficient

What this result cannot establish

• Bounded implementation from supplied acceptance checks and restored CORE definitions; scope choices are explicit below.
• The expression and coefficients are supplied assumptions unless separately validated. Ranges are not probability distributions or confidence intervals.
• Monetary expressions require explicit currency, price basis and reference year. Every monetary term and output must remain safe integer minor units; no monetary conversion is performed.
• Only a declared linear scalar expression is supported. No eval, arbitrary JavaScript or automatic model selection.
• The box range permits independent input extremes, which may not be jointly feasible. All-low and all-high outputs need not be the minimum and maximum.
• An ordinary spreadsheet gives the same arithmetic; no AI uplift is inferred.

What would falsify the assumptions?

• The result loses meaning if coefficients, units, linearity or feasible ranges are wrong.
• Independent measured input/output pairs can contradict the supplied expression.
• This expression does not itself conserve resources; resource constraints need a separate checked model.
• Evaluate the same three substitutions without AI.
• This is deterministic arithmetic over declared assumptions, not correlation or a causal estimate.
• MAR adds risk if users mistake supplied coefficients for evidence; identical manual arithmetic establishes no benefit.
• Assumptions fail if the supplied definitions, units, period, scope or source records are wrong or incomplete.
• An independent source audit or replication may contradict an input or the recorded dependency.
• This method creates no resources. Physical conservation is checked in the inventory and transfer ledgers; descriptive graphs and checklists are not resource-flow models.
• The same equations and review checklist can be reproduced with a spreadsheet or structured human-only process.
• This is deterministic arithmetic, a descriptive comparison or a structured governance check, not a fitted causal effect or forecast. Fictional inputs remain an exercise.
• Equal or worse measured errors, net review effort, disclosure burden or fallback under MAR would show no added benefit or added risk; no improvement is assumed.

{
  "observations": {
    "unit": "staff-hours",
    "valueKind": "quantity",
    "origin": "synthetic",
    "sources": [
      {
        "id": "SYN-001",
        "kind": "synthetic",
        "label": "Fictional linear workload assumptions",
        "url": null
      }
    ],
    "variables": [
      {
        "id": "cases",
        "unit": "cases",
        "low": 10,
        "base": 20,
        "high": 30,
        "justification": "Fictional exercise endpoints; not empirically estimated."
      }
    ],
    "expression": {
      "type": "linear",
      "intercept": 5,
      "interceptUnit": "staff-hours",
      "terms": [
        {
          "variable": "cases",
          "coefficient": 2,
          "coefficientUnit": {
            "numerator": "staff-hours",
            "denominator": "cases"
          }
        }
      ]
    }
  },
  "parameters": {
    "method": "declared-linear",
    "design": "ofat",
    "scenarioCap": 256
  }
}

{
  "lowSetting": 25,
  "baseSetting": 45,
  "highSetting": 65,
  "boxMinimum": 25,
  "boxMaximum": 65
}

Authoritative background references

• U.S. GAO: Cost Estimating and Assessment Guide — Published 12 March 2020. Estimation and sensitivity guidance; no endorsement of this implementation. Reviewed 2026-11-09. Reference, not endorsement or validation.

These sources inform definitions or review practices. They do not validate this implementation or endorse MAR.