https://stefano-blando.github.io/en/tags/factor-models/Factor ModelsHugoBlox Kit (https://hugoblox.com)en-USFri, 06 Mar 2026 00:00:00 +0000https://stefano-blando.github.io/media/icon_hu_8d0dee6c10a3c598.pnghttps://stefano-blando.github.io/en/tags/factor-models/https://stefano-blando.github.io/en/projects/robust-portfolio-optimization/Fri, 06 Mar 2026 00:00:00 +0000https://stefano-blando.github.io/en/projects/robust-portfolio-optimization/<p>This project introduces the <strong>Parallel Factor Space Estimator (PFSE)</strong>, a hybrid framework for robust covariance estimation that resolves a fundamental trade-off in institutional portfolio management: traditional robust estimators (MCD, Tyler) offer strong statistical guarantees but are computationally infeasible for daily rebalancing of 100+ assets, while efficient methods (Ledoit-Wolf, sample covariance) have zero breakdown point against systematic contamination.</p> <p>PFSE exploits a structural insight: during systematic market disruptions — flash crashes, monetary policy shocks, crisis contagion — extreme movements propagate through <strong>common factors</strong>, not idiosyncratic components. By concentrating robust estimation in reduced k-dimensional factor space (k=5 versus p=100–1000), PFSE inherits 25% breakdown point from MCD while achieving <strong>32× computational speedup</strong>.</p> <p><em>Work with Alessio Farcomeni (University of Roma Tor Vergata). Submitted to Computational Economics (Springer), March 2026.</em></p> <h2 id="key-results">Key Results</h2> <p>Validated through three complementary stages:</p> <ul> <li><strong>Monte Carlo (p=100, ε=10% contamination):</strong> PFSE Sharpe 1.42 vs 0.96 for sample covariance — maintains 97% of clean-data performance while sample covariance degrades 31%</li> <li><strong>S&amp;P 500 backtest (2015–2025):</strong> Out-of-sample Sharpe 1.87 vs 1.63 (+14.7%), max drawdown −24.3% vs −34.1% (−29%) during COVID-19, turnover −42%</li> <li><strong>Five stress scenarios:</strong> PFSE rank-1 across all scenarios, average Sharpe 1.67 vs 1.39 (+20%), lowest performance variability (CoV 0.041 vs 0.064)</li> <li><strong>Economic value:</strong> $72M normal-period + $93M stress-period benefits per $1B portfolio, benefit-cost ratio 31:1</li> </ul> <h2 id="interactive-explorer">Interactive Explorer</h2> <p>The interactive app lets you explore the full results: run the PFSE estimation algorithm live, compare methods under increasing contamination levels, examine computational scalability, and inspect S&amp;P 500 cumulative returns across all four market regimes.</p> <p><strong> </strong></p>