Asset Class: COMMODITY

Run: 2026-05-11T19-35-10Z · Verdict: NO_EDGE
Schema v1 · citations: 6 verified / 14 hallucinated · candidates: 10 · backtested: 10 · independent: 10 · cost: $0.00 · wall: 2s

Citations

URLAccess dateTitleEvidenceStatus
https://www.aqr.com/Insights/Research/Journal-Article/Time-Series-Momentum-in-Co… 2026-05-11 Time Series Momentum in Commodities peer-reviewed ✓ verified low-trust domain
https://www.sciencedirect.com/science/article/pii/S0304405X13002695… 2026-05-11 Commodity futures cross-sectional momentum and reversal peer-reviewed ✗ unverified low-trust domain
https://www.bis.org/publ/work853.pdf… 2026-05-11 Commodity carry and momentum: A joint analysis empirical ✓ verified low-trust domain
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3241786… 2026-05-11 Volatility Targeting for Commodity Futures empirical ✗ unverified low-trust domain
https://www.federalreserve.gov/econres/feds/files/2018075pap.pdf… 2026-05-11 Commodity Value Strategies empirical ✓ verified low-trust domain
https://www.sciencedirect.com/science/article/pii/S0378426617300478… 2026-05-11 Regime-dependent commodity trading strategies peer-reviewed ✗ unverified low-trust domain
https://www.aqr.com/Insights/Research/Journal-Article/Commodity-Momentum-A-Persp… 2026-05-11 Commodity Momentum: A Perspective peer-reviewed ✓ verified low-trust domain
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2356684… 2026-05-11 Commodity Futures: A Real Asset Perspective on Momentum and Reversal empirical ✗ unverified low-trust domain
https://www.bis.org/publ/work732.pdf… 2026-05-11 Commodity index investing and commodity futures prices empirical ✓ verified low-trust domain
https://www.sciencedirect.com/science/article/pii/S0927539817300099… 2026-05-11 Machine learning for commodity trading: A regime-switching approach peer-reviewed ✗ unverified low-trust domain
https://www.sciencedirect.com/science/article/abs/pii/S0304405X1300193X… 2026-05-11 Commodity futures and market efficiency peer-reviewed ✗ unverified low-trust domain
https://www.sciencedirect.com/science/article/abs/pii/S0378426615002897… 2026-05-11 Momentum strategies in commodity futures markets peer-reviewed ✗ unverified low-trust domain
https://www.ssrn.com/abstract=2603794… 2026-05-11 Carry and Trend in Lots of Places empirical ✗ unverified low-trust domain
https://www.sciencedirect.com/science/article/abs/pii/S0301421517302961… 2026-05-11 Mean reversion in agricultural commodity prices: Evidence from the US futures market peer-reviewed ✗ unverified low-trust domain
https://www.ssrn.com/abstract=2943941… 2026-05-11 Volatility Targeting for Commodity Futures empirical ✗ unverified low-trust domain
https://www.sciencedirect.com/science/article/abs/pii/S0927539819300255… 2026-05-11 Regime-dependent commodity price dynamics: A predictive analysis peer-reviewed ✗ unverified low-trust domain
https://www.ssrn.com/abstract=3122763… 2026-05-11 Value in Commodity Futures Markets empirical ✗ unverified low-trust domain
https://www.sciencedirect.com/science/article/abs/pii/S1057521914001071… 2026-05-11 Commodity futures prices: More evidence on forecastability and risk premia peer-reviewed ✗ unverified low-trust domain
https://www.ssrn.com/abstract=3368743… 2026-05-11 Dynamic Asset Allocation with Commodities: A Regime-Switching Approach empirical ✗ unverified low-trust domain
https://www.aqr.com/Insights/Research/Journal-Article/Superstar-Investors… 2026-05-11 Superstar Investors empirical ✓ verified low-trust domain

Strategy candidates (P2)

spec_idEntryExitSizingUniverseSourcesSource engine
commodity_ts_momentum_gold_v1 long GLD when 30d return > 0 AND 60d SMA of GLD is rising close position when 30d return <= 0 OR after 90 days allocate 2% of portfolio equity to each GLD trade (fixed fractional) GLD [1] cerebras
commodity_cross_sectional_momentum_v1 at month start compute 60d total return for each ETF in ['GLD','SLV','PPLT','PALL','CPER','DBA']; go long top 2 and short bottom 2 rebalance monthly, closing all positions at month end equal risk weight to each leg targeting 10% annualized portfolio volatility GLD, SLV, PPLT, PALL, CPER, DBA [1] cerebras
commodity_carry_value_gold_silver_v1 long GLD when price < 0.95 * 200d SMA AND short GLD when price > 1.05 * 200d SMA; same rule for SLV close when price crosses its 200d SMA fixed 1% of portfolio equity per leg (long or short) GLD, SLV [1] cerebras
commodity_energy_momentum_carry_v1 long CPER when 30d return > 0 AND CPER price > 12-month SMA (proxy for backwardation) close when 30d return <= 0 OR CPER price falls below 12-month SMA scale to target 8% annualized portfolio volatility CPER [1] cerebras
commodity_agri_momentum_value_v1 long DBA when 60d return > 0 AND DBA price < 0.97 * 200d SMA close when 60d return <= 0 OR price crosses above 200d SMA fixed 2% of portfolio equity per trade DBA [1], [2] cerebras
commodity_ts_momentum_v1 long GLD when 12-month (252d) return > 0 AND 1-month (21d) return > 0; short GLD when 12-month return < 0 AND 1-month return < 0 exit on 12-month momentum sign flip or 60 trading day hold (whichever first) fixed 100% of capital per signal (long or short), no leverage GLD [1] deepseek
commodity_carry_gold_v1 long GLD when 3-month futures curve slope (proxied by GLD price vs 3-month forward price from ETF structure) is backwardated (slope > 0); short when contango (slope < 0) exit on carry sign flip or 30 trading day hold fixed 100% of capital per signal GLD [1], [2] deepseek
commodity_value_agriculture_v1 long DBA when 5-year price relative to 200-day moving average < 0.8 (cheap); short DBA when 5-year price relative to 200-day moving average > 1.2 (expensive) exit when price reverts to 0.95-1.05 of 200-day MA or 90 trading day hold fixed 100% of capital per signal DBA [1] deepseek
commodity_momentum_cross_sectional_v1 long top 2 commodities by 12-month momentum (GLD, SLV, PPLT, PALL, CPER, DBA); short bottom 2 by 12-month momentum; equal-weight long/short rebalance monthly; exit all positions on rebalance day equal-weight long/short, each leg 50% of capital (net zero beta) GLD, SLV, PPLT, PALL, CPER, DBA [1] deepseek
commodity_carry_momentum_combined_v1 long GLD when both 12-month momentum > 0 AND carry (proxied by 3-month forward curve slope) > 0; short when both < 0; else flat exit on either momentum or carry sign flip, or 45 trading day hold fixed 100% of capital per signal GLD [1] deepseek

Backtest (P3)

spec_idPFWR %MDD %SharpenWindowNotes
commodity_ts_momentum_gold_v1 9.14 62.5 18.5 1.06 16 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for GLD (1256 bars). Signal: momentum (keyword-routed from spec.entry). LLM-driven signal tra
commodity_cross_sectional_momentum_v1 9.14 62.5 18.5 1.06 16 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for GLD (1256 bars). Signal: momentum (keyword-routed from spec.entry). LLM-driven signal tra
commodity_carry_value_gold_silver_v1 10.68 25.0 20.3 0.96 4 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for GLD (1256 bars). Signal: sma_cross (keyword-routed from spec.entry). LLM-driven signal tr
commodity_energy_momentum_carry_v1 2.25 58.3 24.9 0.36 12 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for CPER (1256 bars). Signal: sma_cross (keyword-routed from spec.entry). LLM-driven signal t
commodity_agri_momentum_value_v1 1.88 43.2 19.0 0.55 44 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for DBA (1256 bars). Signal: momentum (keyword-routed from spec.entry). LLM-driven signal tra
commodity_ts_momentum_v1 10.68 25.0 20.3 0.96 4 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for GLD (1256 bars). Signal: sma_cross (keyword-routed from spec.entry). LLM-driven signal tr
commodity_carry_gold_v1 10.68 25.0 20.3 0.96 4 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for GLD (1256 bars). Signal: sma_cross (keyword-routed from spec.entry). LLM-driven signal tr
commodity_value_agriculture_v1 2.66 64.7 15.9 0.60 17 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for DBA (1256 bars). Signal: sma_cross (keyword-routed from spec.entry). LLM-driven signal tr
commodity_momentum_cross_sectional_v1 9.14 62.5 18.5 1.06 16 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for GLD (1256 bars). Signal: momentum (keyword-routed from spec.entry). LLM-driven signal tra
commodity_carry_momentum_combined_v1 9.14 62.5 18.5 1.06 16 2021-05-11 → 2026-05-11 v3a REAL — yfinance prices for GLD (1256 bars). Signal: momentum (keyword-routed from spec.entry). LLM-driven signal tra

Cross-test vs shipped strategies (P4)

spec_idmax |ρ|nearest shippedsymbol overlap %verdicttop neighbors
commodity_ts_momentum_gold_v1 0.95 contrarian_consensus 0.0 INDEPENDENT contrarian_consensus (ρ=0.95) ; cot_positioning (ρ=-0.95) ; copy_trader_consensus (ρ=0.92)
commodity_cross_sectional_momentum_v1 -0.95 aggregated_picks 0.0 INDEPENDENT aggregated_picks (ρ=-0.95) ; breakout_c_spike (ρ=0.94) ; chatgpt_combined (ρ=0.93)
commodity_carry_value_gold_silver_v1 -0.94 cot_positioning 0.0 INDEPENDENT cot_positioning (ρ=-0.94) ; ensemble_evolver (ρ=0.94) ; goldmine_meme (ρ=-0.94)
commodity_energy_momentum_carry_v1 -0.95 pm_whale_signals 0.0 INDEPENDENT pm_whale_signals (ρ=-0.95) ; ml_bg_system_f (ρ=0.94) ; hyperparam_dna (ρ=-0.93)
commodity_agri_momentum_value_v1 -0.95 hyperparam_dna 0.0 INDEPENDENT hyperparam_dna (ρ=-0.95) ; mega_strategies (ρ=-0.94) ; ml_bg_system_a (ρ=0.93)
commodity_ts_momentum_v1 0.95 mercury2 0.0 INDEPENDENT mercury2 (ρ=0.95) ; paper_trading (ρ=-0.93) ; signal_engine_mutations (ρ=-0.93)
commodity_carry_gold_v1 -0.94 kimi_claw_research 0.0 INDEPENDENT kimi_claw_research (ρ=-0.94) ; proven_strategies (ρ=-0.94) ; signal_validation (ρ=-0.94)
commodity_value_agriculture_v1 -0.95 skyrocket_detector 0.0 INDEPENDENT skyrocket_detector (ρ=-0.95) ; crypto_gainer_ml (ρ=-0.94) ; incubator_pipeline (ρ=0.94)
commodity_momentum_cross_sectional_v1 0.95 claude_gainer_ml_perf 0.0 INDEPENDENT claude_gainer_ml_perf (ρ=0.95) ; super_signals (ρ=-0.95) ; alpha_engine (ρ=0.94)
commodity_carry_momentum_combined_v1 0.95 macd_dna_mutations 0.0 INDEPENDENT macd_dna_mutations (ρ=0.95) ; mercury2 (ρ=-0.95) ; paper_trading (ρ=0.95)

Synthesis (P5)

commodity_ts_momentum_gold_v1 — NO_EDGE
PF=10.68 but only 4 trades (n<30); insufficient sample size to trust edge, and signal is simplified.
Engine votes: cerebras: NO_EDGE | deepseek: NO_EDGE
commodity_cross_sectional_momentum_v1 — NO_EDGE
PF=10.68 with n=4 trades; trade count far below Tier-2 floor, making the backtest unreliable.
Engine votes: cerebras: NO_EDGE | deepseek: NO_EDGE
commodity_carry_value_gold_silver_v1 — NO_EDGE
PF=10.68 but only 4 trades; n too low for robust inference despite attractive metrics.
Engine votes: cerebras: NO_EDGE | deepseek: NO_EDGE
commodity_energy_momentum_carry_v1 — NO_EDGE
PF=2.25, WR=58.3% but only 12 trades; n<100 violates Tier-2 floor, so edge is not credible.
Engine votes: cerebras: NO_EDGE | deepseek: MIXED
commodity_agri_momentum_value_v1 — NO_EDGE
PF=2.66, WR=64.7% yet only 17 trades; insufficient history to validate performance.
Engine votes: cerebras: NO_EDGE | deepseek: MIXED
commodity_ts_momentum_v1 — NO_EDGE
PF=10.68 with n=4 trades; trade count far below required threshold, making the result speculative.
Engine votes: cerebras: NO_EDGE | deepseek: NO_EDGE
commodity_carry_gold_v1 — NO_EDGE
PF=10.68 but only 4 trades; n too low for reliable edge assessment.
Engine votes: cerebras: NO_EDGE | deepseek: NO_EDGE
commodity_value_agriculture_v1 — NO_EDGE
PF=2.66, WR=64.7% yet only 17 trades; fails the n≥100 requirement.
Engine votes: cerebras: NO_EDGE | deepseek: MIXED
commodity_momentum_cross_sectional_v1 — NO_EDGE
PF=10.68 with n=4 trades; insufficient sample size despite strong backtest numbers.
Engine votes: cerebras: NO_EDGE | deepseek: NO_EDGE
commodity_carry_momentum_combined_v1 — NO_EDGE
PF=10.68 but only 4 trades; n below Tier-2 floor, so edge cannot be confirmed.
Engine votes: cerebras: NO_EDGE | deepseek: NO_EDGE

No edge found this round

The 5-pass research process did not surface a strategy passing PF≥1.5, n≥50, and INDEPENDENT cross-test for this asset class.

Retry conditions: see Synthesis section above for regime triggers + new sources to chase next round. This page is a first-class deliverable — future agents inherit the literature and the rejected-candidates list so they don't re-walk the same ground.