The portal of Helvetia Quant Labs
The portal of Helvetia Quant LabsQuantitative Research & Development
Research methodology
From a testable question to evaluation, review and revision.
Evaluation process
- 01
Define the question
State the hypothesis, assumptions and conditions that would contradict it before evaluating results.
- 02
Design the evaluation
Specify the data, comparison baseline and evaluation periods. Keep data used to develop a hypothesis separate from observations used to test it.
- 03
Measure uncertainty
Consider sample size, dependence between observations, transaction costs and measurement error alongside aggregate results.
- 04
Test sensitivity
Compare periods, market conditions and alternative assumptions. Record where a finding weakens or no longer holds.
- 05
Review and revise
Retain contradictory results and document changes to the hypothesis. A historical result alone does not establish readiness for operational use.