Objective
Improve Varadi Percentile Channels corr vs AS.
Notes: Month-end-only channel hysteresis: corr 0.913→0.992 CAGRΔ +0.59%. Daily-path QuantStrat was the gap. Findings: varadi-percentile-channels-corr.md
Approach
Reuses existing report research/findings/varadi-percentile-channels-corr.md — tables/metrics preserved verbatim via source_report pointer. Body below summarizes grounded evidence from that report and adjacent artifacts.
Files / code / data changed
Associated commits (inferred from log grep t67):
c0917c47research: t67 Varadi Percentile Channels corr 0.913→0.992 (#65)c0917c47research: t67 Varadi Percentile Channels corr 0.913→0.992 (#65)- Source report:
research/findings/varadi-percentile-channels-corr.md - Task-history entry:
research/findings/task-history.json#t67
Results
Key table from source report (preserved):
| variant | corr | CAGRΔ | Apr-2026 ours (AS 7.7%) | notes |
|---|---|---|---|---|
| daily-path hysteresis (t39) | 0.913 | +0.21% | 1.9% | mostly cash+DBC that month |
| PDBC as commodity | 0.908 | +0.33% | — | live UI match; worse hist corr |
| unadjusted Close | ≤0.88 | worse | — | reject |
| abs-norm alternatives / flat=0 | ≤0.91 | — | — | reject |
| month-end-only hysteresis | 0.992 | +0.59% | 8.0% | shipped |
Conclusions and metrics are in research/findings/varadi-percentile-channels-corr.md; see detail page for preserved tables/metrics.
Conclusions
Month-end-only channel hysteresis: corr 0.913→0.992 CAGRΔ +0.59%. Daily-path QuantStrat was the gap. Findings: varadi-percentile-channels-corr.md
Problems / follow-ups
- See source report or PR discussion for follow-ups. No unsupported follow-ups fabricated.
Links
- Task-history:
research/findings/task-history.jsonidt67 - Findings:
research/findings/varadi-percentile-channels-corr.md - Commits:
c0917c47,c0917c47