Does signal memory change when the market gets busy?
A proposed check of the same memory rule across different levels of trading activity.
The question
A fixed window can span different amounts of activity. Would an event-based window and a clock-based window behave differently when evaluated on the same observations?
Study status: proposed, not run. This sample note exists to review the reading experience and research layout. It does not report an observed market result.
Proposed test
Define quiet and busy periods with a rule that uses past activity only. Apply both memory rules to a shared input and record their response paths. Keep the classification rule separate from the outcome being evaluated.
Study record
| Item | Current state |
|---|---|
| Activity classification | Draft, past observations only |
| Clock-based window | To be specified |
| Event-based window | To be specified |
| Comparison results | Not run |
What could mislead us
Changing both the memory definition and the evaluation sample would make the comparison difficult to interpret. Both variants should be evaluated on the same eligible observations.
Results
No results are available. The finished version would show the sample size, excluded observations, comparison with the baseline, uncertainty, and links to reproducible artifacts. Empty result fields are intentional.
Next step
Prepare a small illustrative timeline to explain the two windows. Then document an actual sample before producing any result chart.
Return to the research notebook or explore the Memory series.