THE RESULT
What the data shows
The largest computed metric is 1 for jan; the smallest is 0 for dec. Metric: rate (Zero recorded-area share).
Zero is retained as an observed value. It does not mean that no fire occurred.
| month | events | n | rate | wilson lower | wilson upper |
|---|---|---|---|---|---|
| apr | 5 | 9 | 0.5556 | 0.2666 | 0.8112 |
| aug | 85 | 184 | 0.462 | 0.3914 | 0.534 |
| dec | 0 | 9 | 0 | 0 | 0.2992 |
| feb | 10 | 20 | 0.5 | 0.2993 | 0.7007 |
| jan | 2 | 2 | 1 | 0.3424 | 1 |
| jul | 14 | 32 | 0.4375 | 0.2817 | 0.6067 |
| jun | 9 | 17 | 0.5294 | 0.3096 | 0.7384 |
| mar | 35 | 54 | 0.6481 | 0.5148 | 0.7618 |
| may | 1 | 2 | 0.5 | 0.09453 | 0.9055 |
| nov | 1 | 1 | 1 | 0.2065 | 1 |
| oct | 10 | 15 | 0.6667 | 0.4171 | 0.8482 |
| sep | 75 | 172 | 0.436 | 0.3641 | 0.5108 |
THE METHOD
From source to answer
Zero-area indicator with monthly denominators and Wilson intervals.
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
Confirm the source measurement convention before assigning zeros a physical meaning.
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.