A research strategy becomes a financial commitment when someone allocates capital and depends on its data, orders and controls every day. The final topic connects the programme's historical evidence to that commitment. It asks what a paper system can establish, what requires a limited live trial, and which observations justify retaining, changing or retiring a strategy.
Backtest evaluation provides the starting evidence. It cannot establish that tomorrow's data will arrive before the decision deadline or that a passive order would actually fill. Conversely, an operationally reliable system can implement a losing idea perfectly. Promotion therefore needs separate evidence about correctness, tradability and economic value. A single profitable week cannot settle all three questions.
By the end of this note you should be able to:
- reconcile a frozen research model with prospective signals, orders and positions;
- distinguish replay, shadow operation, paper fills and actual executions;
- write stage-specific promotion and rollback criteria before observing results;
- diagnose data drift, execution deterioration and weakening predictive relationships;
- interpret losses without assuming every drawdown means model failure;
- assemble a research dossier whose decisions can be independently reproduced.
In this lesson
- Contents
- The big picture
The full lesson is for members.
Read the full lessonNew to SelfMadeQuant.com? See membership options or browse the Quantitative Research, Data and Backtesting syllabus.