Uncertainty ranges benchmarks
19 of 19 cases match Python Prophet.
Both libraries simulate possible futures with random draws, so individual simulated values never match. Each range must still match Python's within the variation expected from that many draws, on every trend type.
Each case runs the same input through Effect Prophet and Python Prophet 1.4.0, then compares every forecast row, the trend, each component and the fitted noise. A case matches when all of them agree within its tolerances, and only matching cases are timed.
Each case's dot shows its result: matches, matches with internal differences flagged, intentional difference, doesn't match. Hover over a dot for details.
How to read the tables:
- Effect output and Python output are each library's final forecast value. Open a case to see every row side by side.
- Largest difference is the biggest gap between the two forecasts, as a share of the input data's range.
- Effect time and Python time are the median time for the timed step, usually fitting and predicting. Open a case for every step, including first-run and save/load times.
- Flagged cases match, but the two libraries stopped their fitting search at slightly different points. The case page shows the internal numbers.
About these runs
Each trend's cases come from that trend's own run.
Linear growth
- Run
2026-10-20T170110-379Z-27b7ec98on 2026-10-20, commit27b7ec9 - Both libraries ran in
linux/arm64containers on Apple M4 Pro - effect-prophet 0.0.0
- effect 4.0.0
- prophet 1.4.0
- Full benchmark report
Flat growth
- Run
2026-10-20T171807-628Z-27b7ec98on 2026-10-20, commit27b7ec9 - Both libraries ran in
linux/arm64containers on Apple M4 Pro - effect-prophet 0.0.0
- effect 4.0.0
- prophet 1.4.0
- Full benchmark report
Logistic growth
- Run
2026-10-20T174321-690Z-27b7ec98on 2026-10-20, commit27b7ec9 - Both libraries ran in
linux/arm64containers on Apple M4 Pro - effect-prophet 0.0.0
- effect 4.0.0
- prophet 1.4.0
- Full benchmark report
Timings come from one machine and describe this workload only. They are not a general speed ranking. Where Python is faster, it is often because Python runs Stan's optimizer as compiled C++, while Effect Prophet runs the same algorithm as WebAssembly with narrower vector instructions. The gap is widest for Newton's method on short histories with many parameters.
Linear growth
Both libraries simulate possible futures with random draws, so individual simulated values never match. Each range must still match Python's within the variation expected from that many draws. All linear growth benchmarks →
| Case | Input | Effect output | Python output | Largest difference | Timed step | Effect time | Python time |
|---|---|---|---|---|---|---|---|
| Defaults, three years of ranges | 1095 rows | 611.2687 | 611.3156 | 0.011% | Simulate uncertainty | 71.3 ms | 379 ms |
| Defaults, ranges on irregular dates | 192 rows | 108.5108 | 108.4320 | 0.15% | Simulate uncertainty | 11.9 ms | 158 ms |
| Samples with repeated timestamps | 96 rows | 17.9039 | 17.9039 | 0.0013% | Simulate uncertainty | 498 µs | 16.9 ms |
| Ranges over past dates | 96 rows | 110.0950 | 110.0950 | < 0.0001% | Simulate uncertainty | 327 µs | 20.3 ms |
| Samples for one future date | 96 rows | 110.0819 | 110.0819 | < 0.0001% | Simulate uncertainty | 208 µs | 17.2 ms |
| Ranges with mixed effects | 96 rows | 45.1353 | 45.1353 | < 0.0001% | Simulate uncertainty | 398 µs | 26.0 ms |
| Samples with mixed effects | 96 rows | 45.1353 | 45.1353 | < 0.0001% | Simulate uncertainty | 726 µs | 68.6 ms |
Flat growth
Both libraries simulate possible futures with random draws, so individual simulated values never match. Each range must still match Python's within the variation expected from that many draws. All flat growth benchmarks →
| Case | Input | Effect output | Python output | Largest difference | Timed step | Effect time | Python time |
|---|---|---|---|---|---|---|---|
| Country calendar uncertainty replay | 96 rows | 21.4447 | 21.4446 | 0.0048% | Simulate uncertainty | 318 µs | 14.8 ms |
| Defaults, ranges with a weekly pattern | 96 rows | 35.0025 | 35.0025 | < 0.0001% | Simulate uncertainty | 4.48 ms | 125 ms |
| Ranges with mixed effects | 96 rows | 35.0283 | 35.0283 | < 0.0001% | Simulate uncertainty | 1.10 ms | 73.6 ms |
| Samples, eight months | 256 rows | 19.5305 | 19.5305 | < 0.0001% | Simulate uncertainty | 1.21 ms | 22.3 ms |
Logistic growth
Both libraries simulate possible futures with random draws, so individual simulated values never match. Each range must still match Python's within the variation expected from that many draws. All logistic growth benchmarks →
| Case | Input | Effect output | Python output | Largest difference | Timed step | Effect time | Python time |
|---|---|---|---|---|---|---|---|
| Defaults, ranges with changing growth | 365 rows | 193.9557 | 193.9523 | 0.0086% | Simulate uncertainty | 23.5 ms | 282 ms |
| Defaults, ranges with changing growth, second seed | 365 rows | 193.9557 | 193.9523 | 0.0086% | Simulate uncertainty | 23.1 ms | 288 ms |
| Samples, weekly pattern | 96 rows | 71.9060 | 71.9060 | < 0.0001% | Simulate uncertainty | 1.22 ms | 80.7 ms |
| Ranges, eight months | 256 rows | 98.0804 | 98.0804 | < 0.0001% | Simulate uncertainty | 861 µs | 26.6 ms |
| Ranges with an explicit floor | 96 rows | 72.2190 | 72.2190 | < 0.0001% | Simulate uncertainty | 945 µs | 89.0 ms |
| Samples with an explicit floor | 96 rows | 72.2190 | 72.2190 | < 0.0001% | Simulate uncertainty | 563 µs | 26.0 ms |
| Ranges with mixed effects | 96 rows | 72.2270 | 72.2270 | < 0.0001% | Simulate uncertainty | 629 µs | 34.6 ms |
| Samples with mixed effects | 96 rows | 72.2270 | 72.2270 | < 0.0001% | Simulate uncertainty | 794 µs | 27.5 ms |