Regression-weighted butterfly
A systematic fixed income approach—Regression-weighted butterfly—defined by explicit rules, testable on history, and fragile when costs or regimes change.
Overview
Regression-weighted butterfly sits in the Fixed Income chapter of the systematic catalog. On QUSXFI we treat it as a testable hypothesis: specify entries, exits, sizing, and costs—then ask whether edge survives out-of-sample scrutiny.
Discretionary traders often arrive at similar ideas intuitively; the quantitative version forces you to write the rule before you see the next bar. That discipline is what makes results reproducible—or exposes them as luck.
Based on the research catalog 151 Trading Strategies (Kakushadze & Serur, 2018), section 5.8. Educational summary—not a replication of the full formal definition.
Multi-Leg Payoff Logic
Pin and spot-vol interaction near expiry can turn Regression-weighted butterfly from 'defined risk' into gamma you did not model.
Map every input Regression-weighted butterfly needs in Fixed Income—prices, vol surfaces, fundamentals, or legal milestones—and verify point-in-time integrity.
Implementation and Research Process
Commission-scale Regression-weighted butterfly honestly; multi-leg edges often die net of costs.
Decompose Regression-weighted butterfly into signal, portfolio construction, and execution modules—each must be path-independent given the same historical tape.
Document Regression-weighted butterfly capacity in Fixed Income: intended participation versus average daily volume.
Risk: What Breaks This Strategy
Multi-leg structures (Regression-weighted butterfly) multiply commission, margin, and operational error. One leg fills, another does not—you are suddenly naked risk.
Small moves in spot and vol interact nonlinearly; a 'defined risk' label does not mean defined stress behavior.
Adjustments mid-trade often become discretionary—exactly what systematic rules tried to avoid.
Common Mistakes to Avoid
- Using academic §5.8 definitions for Regression-weighted butterfly while ignoring borrow, margin, or contract specs.
- Under-budgeting commission and slippage on Regression-weighted butterfly multi-leg packages.
- Adjusting Regression-weighted butterfly mid-trade without pre-written rules—discretion destroys the systematic label.
- Stacking Regression-weighted butterfly with correlated sidebar strategies without netting exposures.
How to Study This Strategy
- Run a paper book on Regression-weighted butterfly for a full signal cycle; export trades and tag regimes manually.
- Compare Regression-weighted butterfly to one sidebar alternative net of costs—document why you chose this structure.
- Restate Regression-weighted butterfly (§5.8) as numbered rules another researcher could implement cold.
- List every data field Regression-weighted butterfly needs in Fixed Income; verify point-in-time integrity.
- Map Regression-weighted butterfly to Basic Trading chart concepts you will use as filters—not as substitutes for rules.
Key Takeaways
- Regression-weighted butterfly multiplies legs, margins, and operational failure modes—one missed fill creates naked exposure.
- Document adjustment rules for Regression-weighted butterfly in advance; mid-trade discretion destroys systematic claims.
- Commission and slippage scale with leg count—net edge often lives or dies on costs.
- Small spot-vol moves interact nonlinearly; stress jointly, not one greek at a time.
- Paper-trade Regression-weighted butterfly with full leg fills simulated at bid/ask before debating live capital.
Learning Tip
Explain Regression-weighted butterfly to someone who only knows Basic Trading charts—if you need unexplained jargon, the spec is not ready.
Explore related strategies in the sidebar or return to the full catalog.