What a model card is forThis page exists so any Warconomy forecast produced by this model can be audited: what it assumes, where it breaks, how its uncertainty is built, and what simple baseline it must beat to be worth running. Forecasts are decision-support, not advice.
At a glance
| Model ID | trend-continuation-v1 |
|---|---|
| Version | 1.0 |
| Method family | rule based projection |
| Intended use | Project a continuing series forward by extending its recently observed rate of change, for metrics that move gradually and are published on a regular cadence. |
| Update cadence | monthly |
| Analyst judgment | Yes — a human sets or adjusts key parameters (see below) |
| Forecasts using it | 5 |
Assumptions
- The most recently observed rate of change persists over the forecast horizon.
- No discrete shock (a major offensive, a ceasefire, a policy reversal) materially breaks the trend within the horizon.
- The publishing institution keeps its measurement definition unchanged over the horizon.
Known weaknesses and failure conditions
- Cannot anticipate discrete shocks — precisely the events that matter most in conflict economics.
- Uncertainty widens rapidly with horizon length; beyond roughly two publication cycles the interval becomes too wide to be decision-useful.
- No seasonality adjustment; series with strong seasonal structure need a different method.
- Interval width is set by analyst judgment anchored to historical revision size, not derived from a fitted error distribution, because most of these series have too few comparable published points to fit one.
How uncertainty is built
A plausible-range bracket, widened for longer horizons, reflecting the compounding risk that the constant-rate assumption breaks down. Not a statistical confidence interval — labeled as plausible-range throughout.
Validation and baseline comparison
Compared against a no-change naive baseline on historical published points where two or more comparable rounds exist. See /forecast-track-record for the hindcast demonstrations.
Model change history
| Version | Change |
|---|---|
| 1.0 | Initial production release of this model card, issued with Warconomy's first forecast portfolio. |
Model versions are immutable. A material methodology change produces a new version, and forecasts issued under the old version keep pointing at the version that produced them.
Forecasts produced by this model
- Sudan IDP count stays above 8 million in IOM DTM's next published round — active
- OHCHR reports at least one month in H2 2026 above the H1 2026 monthly average — active
- The next Ukraine RDNA raises the ten-year reconstruction estimate above $588 billion — active
- Sudan IDPs by December 2026 (superseded) — superseded
- Hindcast: would trend continuation have beaten 'no change' on Sudan's IDP total? (hindcast demonstration)
Related
All forecasting methods · Responsible forecasting policy · Active forecasts · Track record