VTI/SPY 10-DMA × VIX edge parameter sweep (qm-wr6.1)
Shipped-as-strategy qm-wr6.1 ·
Decision
Ship cashonupper + meanortime (edgefreq=0.02, meanrevert=0.5%, timeexit=5d) with nextclose execution on SPY + ^VIX.
Provenance
Equity vs S&P (aligned, rebased to 1.0)
No signal overlay for this experiment (parameter sweep / search)
Aligned window, both rebased to 1.0 at start — Aligned window, both rebased to 1.0 at start for fair comparison · S&P dashed, experiment solid. X = Date, Y = Equity (rebased to 1.0).
Signal driver & thresholds (x-aligned)
VTI/SPY × VIX edge parameter sweep
Edge frequency 0.02, mean-revert 0.5%, time exit 5d — see payload for full grid.
Stats
| CAGR | 10.78% | Sharpe | 64.43% |
|---|---|---|---|
| Sortino | 102.67% | maxDD | -55.19% |
Disposition
Ship **cash_on_upper** + **mean_or_time** (edge_freq=0.02, mean_revert=0.5%, time_exit=5d) with **next_close** execution on SPY + `^VIX`. Engine is long-only: upper-edge "short" is residual T-bill cash. SPY proxies VTI (VTI was absent from `market.sqlite` at research time). Full-sample Sharpe/CAGR stay slightly **below** raw SPY buy&hold (Sharpe≈0.63 vs 0.64; CAGR≈10.3% vs 10.8%). This is an investable defensive overlay with researched defaults, not a claimed SPY-beating alpha product. Opportunistic `both_edges` / `long_on_lower` shells score higher Sharpe only when barely invested (~1–10%); ranking penalizes those for catalog use.
Follow-on work
- Follow-up qm-wr6.1
- Follow-up qm-wr6.1.json
- Follow-up qm-wr6.2
Errors & data gaps
No additional vendor or coverage gaps beyond frozen panel window; see finding for panel dates and synthetic reconstruction limits.
Follow-up
gh issue create --title "Follow-up VTISPY 10-DMA VIX edge parameter sweep qm-wr61" --body "Follow-up to qm-wr6.1: Decision" --label "priority:p2" --label "pipeline:research" --label "agent:ready"