Oil–Rates Monitor

Active Macro Regime

—

Oil → BEI Correlation — vs — full sample Brent close, daily, r

Breakeven Channel Share — vs — full sample % of total oil→yield effect

Channel Status — Based on current r vs thresholds

Oil Price Brent / WTI crude

r = —

▶

Inflation expectations repricing

Breakeven 10Y Market inflation expectations

— of effect

▶

Dominant transmission path

Nominal 10Y TY futures yield

Empirical result 2008–present: when oil moves, ~— of the resulting nominal yield change is explained by breakeven repricing, not real yields. Real 10Y correlations with oil are near zero. Full-sample Brent→breakeven r = —; current regime r = —.

Regime Comparison · Brent

Regime	Oil→BEI r	BEI channel

20-Day TY Position · Consensus

—

⚠ This is a 20-trading-day rates-risk overlay, not a directional trading instruction. Do not override upcoming CPI, NFP, Fed decisions, auction risk, or desk flow.

OOS-Weighted 20d 10Y Move — —

Positive = yield rises → Short TY
Negative = yield falls → Long TY

Per-Factor-Family Signals · 20-day horizon

Linear vs Nonlinear Model Comparison

walk-forward OOS, 20d horizon

OLS is the linear baseline. Regime interaction is still OLS, but lets the current macro regime change the factor beta. Random forest is the nonlinear challenger.

Model Family	Tested	Positive OOS	Best OOS R2	Best Validated Sharpe	Current Read

Regime Glossary

macro state, not politics

GFCGlobal Financial Crisis, 2008-2009 stress period.

ZIRP / QEZero-rate and quantitative-easing period after the crisis.

NormalizationFed hikes / balance-sheet normalization before COVID.

COVID2020-2021 pandemic shock and policy response.

Inflation ShockHigh inflation, supply shock, aggressive Fed hiking.

RecentPost-inflation-shock current market state.

All Tested Data Inputs — Signal Filter Audit

Input Factor	Family	OOS R²	Dir. Acc.	20d Forecast	Status	Why included / excluded

Best Strategy Equity Curve

BCOM (broad commodity basket) 20d signal → TY futures long/short. Rule: predicted 10Y yield ↑ → short TY; predicted yield ↓ → long TY. 5 bp signal threshold. No transaction cost shown; see leaderboard for cost-adjusted results.

Best Validated Backtest

TY futures, 2012–2026

Backtest period starts after the model has seen ≥1,000 days of training data. Sharpe = annualized return ÷ annualized vol (daily PnL).

Strategy Leaderboard

sorted by Sharpe - diagnostic, includes negative OOS R2

Strategy	Family	Sharpe	Total Return	Max DD	Hit Rate	Trades	OOS R²