Research

Authority transitions leave tracks in the public record.

That is the premise, and it might be wrong. The research program is how we find out — estimates for named regulatory, guideline, and coverage decisions, registered before the outcome is known and then checked against what happened. The method is public so the answer does not depend on trusting us.

Research questions

Can defined public medical transitions be forecast prospectively?

in progress

Do the numbers mean what they say — does a 70% forecast happen about seventy times in a hundred?

planned

Does NextConsensus outperform base-rate, heuristic, expert, and LLM baselines?

planned

Which evidence and institutional features improve forecasts?

planned

How early can a forecast be useful before the false alarms cost more than the warning is worth?

planned

Which target rungs are forecastable — and which are not?

planned

How many resolved forecasts does one kind of decision need before its record means anything?

planned

The transition model

Evidence, recognition, and authority action move on different clocks. The model estimates the probability of a defined rung 5–7 transition by a deadline from the upstream states observable at registration.

Research program

The research program determines whether forecasting is possible before any enterprise application is attempted. Public research tests the method. Enterprise pilots test utility.

Retrospective backtesting

Reconstruct forecasts from historical evidence states. Hold out future information. Score against outcomes. Control for temporal leakage.

Prospective registration

Register forecasts before outcomes are known. Freeze evidence state. Publish registration timestamps. Adjudicate outcomes independently.

Calibration analysis

Compute Brier scores, reliability curves, and baseline comparisons per cohort, per event class, and per method version.

Boundary determination

Identify which target rungs are forecastable and which are not. Report tier verdicts: detection, nowcasting, forecasting, or unsupported.

Boundary

Public research validates the forecasting method. Enterprise applications validate forecast utility in decision workflows. These are separate programs with separate evaluation criteria.

NC's forecasting methods are developed and evaluated through public research before being applied to customer contexts.

Research outputs

Registered forecast specifications with defined resolution rules
Retrospective backtest reconstructions with temporal leakage controls
Prospective forecast cohorts with independent adjudication
Calibration analyses per event class and authority
Baseline comparison results
Reproducible tools via Refract Refract on GitHub Refract Licensing