The research question
Does persistence add information beyond the exposure taken to earn it?
A study of market behaviour following unusually large price moves, where timestamp quality and event definition matter as much as the predictor.
The technical work
Signal construction across horizons, volatility scaling, rebalancing and benchmark comparison.
The challenge is to connect the economic idea to a precise, reviewable experiment. Data transformations, the moment of a decision and the interpretation of the resulting return series all belong to that specification.
What to challenge
These are evaluation considerations for this domain, not a checklist presented as already passed.
- Exposure-matched benchmarks
- Parameter stability through changing regimes
- Turnover, roll and financing assumptions
- Holdout performance not used for model selection
The data boundary
Adjusted asset histories or investable futures returns, plus the rates and costs required by the chosen instruments.
Descriptions are intentionally high-level. Parameters, execution rules and non-public source data are not published here. Withholding an implementation is different from overstating its evidence.
The interesting question is not whether a backtest looks good. It is what would have to be true for the result to mean something.
For the broader framework, see validation. For the engineering context, see the research stack.