THE RESULT
What the data shows
The largest computed metric is 1 for 1.0; the smallest is 0.4104 for 0.01. Metric: area_share (Share of recorded hectares).
A small number of large outcomes can dominate mean error; records are not guaranteed independent fire events.
| top record fraction | records | area share |
|---|---|---|
| 0.01 | 6 | 0.4104 |
| 0.05 | 26 | 0.6715 |
| 0.1 | 52 | 0.8051 |
| 0.2 | 104 | 0.914 |
| 0.5 | 259 | 0.9994 |
| 1 | 517 | 1 |
THE METHOD
From source to answer
Cumulative area shares at fixed top-observation fractions.
The Python source exposes this study’s transformations. The complete project download includes shared preparation and evaluation routines.
THE NEXT DECISION
What follows from the finding
Use the heavy-tailed outcome to choose regression metrics and uncertainty checks.
Where the conclusion stops
The sample consists of recorded fire observations, not all places and times at risk. Zero recorded area is not evidence of no ignition. No precise dates or event identifiers are supplied. This is not an emergency forecasting system.
Related studies may reuse observations or holdouts. These are historical analyses; associations and backtests do not demonstrate commercial impact. Further model tuning needs new, untouched evaluation data.