The research question
When is a relationship economically meaningful rather than a coincidence in the sample?
Relative-value research into two share classes of the same issuer, including liquidity, voting rights and the practical cost of holding both legs.
The technical work
Spread construction, statistical relationship tests, rolling formation windows and portfolio-level implementation.
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.
- Stability after the formation period
- Both hedge legs aligned by timestamp
- Transaction costs, financing and borrow where relevant
- Selection of pairs kept separate from evaluation
The data boundary
Price histories for both legs, corporate-action or contract adjustments where applicable, and defensible execution-cost estimates.
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.