pairs trading
Among 12 exchanges we tested, Japan (JPX) is the only market with a positive Sharpe ratio (+0.141) and the highest investment rate: 95% of years had active pairs. We dig into why Japan's market structure makes it uniquely suited for pairs trading.
pairs trading
We screened 3,701 US large-cap stocks for pairs trading candidates: 884,000 same-sector correlation tests, 2,579 candidates surviving all filters. 87% came from Financial Services. Here's why, and what the data looks like.
pairs trading
Pairs trading bets that two related stocks will snap back together after diverging. The theory is sound. Gatev et al. (2006) found 11% excess returns from 1962-2002. We ran the same approach on 2005-2024 across 12 exchanges. Here's what we found.
small cap
We backtested small-cap value on LSE from 2000 to 2025. 15.44% CAGR with a max drawdown of just -10.89%. 0.750 Sharpe ratio, +14.21% excess over the FTSE 100. The best risk-adjusted result of any exchange in our 14-market comparison.
value investing
We ran Graham's net-net strategy on 7 global markets over 24 years with local benchmarks. Japan beat the Nikkei by +3.3% per year. Hong Kong lost 86% peak-to-trough. India badly trailed the Sensex. Here's the full cross-market comparison.
roic
Sustained ROIC on SHH+SHZ from 2000 to 2025. 4.54% CAGR vs 7.30% SPY benchmark. Quality spread still works (+1.67% vs low ROIC) but state influence and retail sentiment dilute the signal.
roic
Sustained ROIC on HKSE from 2000 to 2025. -0.48% CAGR vs 7.30% SPY. Negative absolute returns, but the persistence premium (+1.76%) still separates sustained from single-year ROIC.
roic
We backtested sustained ROIC on 13 exchanges worldwide. India leads at 15.55% CAGR (+8.25% alpha). Clear geographic divide between emerging markets where quality works and Asian markets where it breaks down.
backtests
We tracked 30 major US corporate spinoffs (2011-2024) at six time horizons. The forced-selling effect is real and statistically confirmed at T+5 (-3.5%, p<0.05). The long-run recovery is positive but driven by outliers — GEV +108%, REZI -68%. Full data and SQL included.
analyst ratings
We tested analyst upgrade clusters on US exchanges from 2019-2025. When 2+ analysts move bullish simultaneously, the T+1 abnormal return is +0.475% (t=11.42). But the real finding is in the cluster size: medium clusters (delta 3-4) produce +1.068% at T+21, while large clusters fade completely.
margin expansion
We tested margin expansion as a stock signal across all US exchanges from 2000 to 2025. The surprise: stable margins (9.72% CAGR) beat expanding (6.12%) and contracting (5.29%). $10,000 in stable-margin stocks grew to $101,587 vs $44,188 for expanding margins.
margin expansion
We tested margin expansion on BSE and NSE from 2000 to 2025. Contracting margins returned 17.65% CAGR vs 3.41% for expanding. The quality signal runs backward in India. Cyclical sectors and growth dynamics flip the expected relationship.