End-to-End Python implementation of STRAPSim: a novel portfolio similarity metric from Li et al. (2025). Combines Random Forest proximity learning with residual-aware bipartite matching to quantify economic substitutability between ETF baskets. Full replication pipeline included.

📊 Calculate portfolio similarity metrics to enhance ETF alignment and optimize trading strategies in quantitative finance.

The credit API for AI agents

Is AI displacement risk priced in corporate bonds? Firm-level exposure (industry automatability times labor intensity) on monthly Z-spreads for 344 issuers. A clean, well-identified null.

ML for endogenous OTC bond markets, where a dealer's policy reshapes future flow. Reframes learning as finding a fixed-point equilibrium stable under self-induced market adaptation.

End-to-end Python implementation of Dickerson, Mueller & Robotti (JFE 2023). Implements Dick-Nielsen TRACE cleaning, KRS misspecification-robust two-pass CSR, BKRS jackknife bias-corrected Sharpe ratios, and Fama-MacBeth regressions to rigorously identify priced risk factors in U.S. corporate bonds. Prevents false discoveries.

Code for the MSc dissertation “Fair Value, Residual Convergence and Dislocation Signals in US Bank Corporate Bonds”.