2024-12-30: New Year, New Curve - Relative Value to Jumpstart 2025
Initiate a 2y fwd 1s2s15s PC1 hedged belly payer position (fly weights: -0.92, 2, -1.54) at -199.868 bps. ~2 week holding period. Base case: -187.778 bps, Bull case: -176.685 bps, Stop-loss: -198.872 bps, 1M Carry and Roll: -12.20 bps, 3M Carry and Roll: -13.09 bps
2024-12-20: Ahead of the Curve: Forward-Looking Hedge Ratios for Cash Treasuries
Abstract: This paper introduces a framework for deriving forward-looking hedge ratios for curve and fly trades in the US Treasury (UST) market by leveraging the information embedded in the yield curve spread option (YCSO) and swaption markets respectively. Traditional approaches to hedging directional interest rate exposure often rely on historical correlations, static yield curve relationships, and/or first order approximations which can lag dynamic market conditions and specifically are prone to failure during periods of macro regime shifts. By incorporating the implied volatility of swaptions and implied correlation from the YCSO market, we can construct the covariance structure and perform every fixed income professionals’ favorite linear algebra tool: Principal Components Analysis. We will first give a brief refresher of PCA in the rates market then an example of “standard” backward-looking way to calculate hedge ratios for yield curve trades. We then provide a quick primer on YCSOs and figures of deriving these hedge ratios from market quotes obtained from J.P. Morgan Markets. We will finish with a discussion of the holes in this framework and the instability of eignvectors over time
PCA applied on the interest rate term structure is not a novel idea and is required reading for any first-year analyst on a rates desk. Where I want to focus this report is building an intuitive and geometric understanding of PCA applied to the US Treasury Market through interactive visualization powered by Plotly. I will provide shallow background needed to grasp how traders can apply PCA to USTs as a tool to set up/screen for Relative Value (RV) and also derive hedge ratios for curves and flies. For a deeper understanding of Fixed Income, I point you to Tuckman. For a better mathematical treatment of PCA (at the undergrad level), I point you to Strang or ESL.
I wanted to see if I could write a couple of Python scripts to reproduce a Rates RV Trade idea that Jay Barry & Co. at JPM wrote up. Turns out I can - numbers are a bit off (couple of bps) but thats from me using FedInvest EOD markings. My charts are prettier!