Building CetaResearch. Data analysis for policy, research, and startups.
small cap
We backtested a small-cap value screen on US stocks from 2000 to 2025. Buy the 30 cheapest small caps by P/B with quality controls. Result: 18.10% CAGR vs 7.85% for SPY, with half the drawdown.
small cap
We backtested small-cap value on the Johannesburg Stock Exchange (JSE) from 2000 to 2025. 12.80% CAGR, +4.95% excess over SPY. Strong mid-2000s commodity boom returns, but the 0.163 Sharpe ratio reveals the cost of South Africa's high risk-free rate.
earnings
Taiwan is the clearest exception in a global beat-and-dip study. 410 events from 2014–2025 show statistically significant mean reversion: +2.23%** at T+21 overall, and +7.35%** for the 10–20% dip category. Most other markets show flat or negative drift.
sector rotation
Sector mean reversion on SIX large caps from 2000 to 2025: 4.69% CAGR in CHF vs 2.31% for the SMI, a +2.38% annual edge. Financial Services is the most-selected sector at 33% of quarters. Against SPY (8.02%) the strategy still trails by -3.33% a year.
backtests
We ran the yield gap strategy on Stockholm Stock Exchange stocks from 2000 to 2025. The result: 12.21% CAGR vs 2.95% for OMX Stockholm 30 (+9.26% excess), the highest return of the 20 markets we tested, with an 80% annual win rate against the local index.
earnings
We tested 13,098 beat-and-dip events on US stocks from 2000–2025. Stocks that beat earnings but sell off 5%+ don't recover. They keep underperforming. T+63 CAR is -0.72% (t=-4.03). The sell-off is not an overreaction; it's information.
equity growth
We backtested the Compounding Equity Screen on TSX stocks from 2000 to 2025. 6.02% CAGR vs 3.95% for the TSX Composite, a +2.07% annual excess. Strong outperformance in the first decade and in recovery years (2009, 2010, 2020), with a severe 2008 drawdown of -40.6%.
sector rotation
Sector mean reversion on HKSE large caps from 2000 to 2025: 6.87% CAGR vs 1.61% for the Hang Seng, a +5.26% annual outperformance. 2009 was +77.76%, 2008 was -47.51%. Sharpe 0.148 against a Hang Seng that scored negative at -0.065. Extreme year-to-year variance.
etf
We backtested a portfolio of the 30 least ETF-owned quality stocks on Indian markets (BSE + NSE) over 20 years. Result: 4.47% CAGR vs 10.61% for SPY, but a 50% win rate and +35.7% excess return in 2008. India's concentrated ETF market creates real pockets of under-coverage.
etf
We backtested a portfolio of the 30 least ETF-owned quality A-shares over 20 years. Result: 8.75% CAGR vs 10.61% for SPY, the closest to matching the benchmark among strong-signal exchanges. But two Chinese bull markets (2006, 2014) account for most of the outperformance.
etf
We backtested a portfolio of the 30 least ETF-owned quality stocks on XETRA over 20 years. Result: 7.12% CAGR vs 10.61% for SPY, but the best Sharpe ratio (0.260) among all strong-signal exchanges. Germany's Mittelstand offers the most efficient anti-crowding returns.
working capital
We tested working capital efficiency as a stock selection signal across all US exchanges from 2000 to 2025. The result: 11.07% CAGR vs 7.64% for the S&P 500, with a down capture ratio of just 13%. The portfolio lost 13 cents for every dollar the market lost in downturns.