Source first
Each study identifies its price source, publication date, period, and whether prices are synthetic.
Investment research, with the evidence in view
See the whole argument.
Explore a strategy, inspect how it was tested, and see where the result breaks. Every number leads back to a dated study artifact.
01 / Research workspace
Honest research starts with a baseline and ends with a failure mode.
Read only · Simulation ceiling L2The research desk
Study explorer
Published experiments, preserved with their source data and assumptions.
The study could not be loaded.
Selected experiment
Selected CAGRAnnualized, out of sample
Baseline CAGRSame validation windows
Maximum drawdownSelected strategy
Backtestable shareOf the full allocation
Simulation path
Value of one unit
Illustrative backtest path. It is distinct from the out of sample validation metrics above.
Walk forward validation
The verdict
- Validation folds
- Selection metric
- Source
- Published
Against alternatives
Full-period leaderboard
Swipe the table for Sharpe and drawdown.
| Theory | CAGR | Sharpe | Drawdown |
|---|
These in-sample rankings use the full study period. The walk-forward verdict above uses separate test windows.
Audit trail
What this covers
Instruments
Assumptions for allocation target
Open the complete study JSON How to read this
A result should show its limits.
Compare fairly
The passive baseline and later, untouched validation windows put a compelling training score in context.
Stay in control
This workspace reads research artifacts. It does not connect accounts, draft orders, or move money.