Reproduce the sample
This is a computed run on synthetic data, with seed 42, 160 symbols and data from 2015-01-01 to 2024-12-31. The backtest covers 2019-01-01 to 2024-12-31. All README figures and this report come from the same pipeline state.
Open the full report · Run manifest · Daily results CSV · Configuration · Metrics JSON
Recorded performance
Values below are decimal ratios unless named otherwise; for example, 0.01 CAGR is 1%. Sharpe, Sortino, Calmar and information ratio are ratios, not percentages.
| Metric | Value |
|---|---|
| start | 2019-01-01 |
| end | 2024-12-31 |
| n_periods | 1566 |
| years | 6.21429 |
| periods_per_year | 252 |
| total_return | 0.0520076 |
| cagr | 0.00819204 |
| ann_vol | 0.10748 |
| ann_downside_dev | 0.0758553 |
| sharpe | 0.12962 |
| sortino | 0.183659 |
| calmar | 0.0429131 |
| max_drawdown | -0.190898 |
| max_drawdown_date | 2020-12-31 |
| max_drawdown_duration_days | 1065 |
| var_95 | -0.0109244 |
| cvar_95 | -0.01643 |
| skew | -0.00136024 |
| kurtosis | 3.43624 |
| best_period | 0.0375985 |
| worst_period | -0.0349239 |
| positive_period_share | 0.427842 |
| avg_period_return | 5.52839e-05 |
| hit_rate | 0.427842 |
| avg_turnover | 0.191267 |
| cost_drag_ann | 0.00144403 |
| benchmark_return | -0.267025 |
| benchmark_cagr | -0.0487597 |
| benchmark_vol | 0.241829 |
| active_return | 0.319032 |
| beta | 0.371263 |
| alpha_ann | 0.0216326 |
| correlation | 0.835341 |
| r_squared | 0.697795 |
| tracking_error | 0.163123 |
| information_ratio | 0.212566 |
| up_capture | 0.388454 |
| down_capture | 0.37303 |
| treynor | 0.0375247 |
| gross_total_return | 0.0615388 |
| gross_cagr | 0.00965637 |
| gross_ann_vol | 0.107485 |
| gross_sharpe | 0.143122 |
| gross_sortino | 0.202939 |
| gross_calmar | 0.0508307 |
| gross_max_drawdown | -0.189971 |
| cost_drag_cagr | 0.00146433 |
| gross_benchmark_cagr | -0.0487597 |
Run it again
pip install -e ".[viz]"
python scripts/publish_sample.py
This command reads configs/default.yaml directly, without environment overrides,
and refuses a non-synthetic provider. Review the manifest's Python and dependency
versions to reproduce this environment. Exact source file hashes accompany the
source revision.
Output files go to docs/sample/; README figures go to assets/.
For normal research runs with other providers or settings, use the Quickstart.
Read the validation limits before interpreting performance.