Research
Open replications of published trading-strategy research — rules stated precisely enough to rebuild and verify, not academic summaries.
For each study we take a paper's claimed edge, codify the rules in readable code, backtest it on real data, and report honestly whether it held up — including where our data or methods fall short. You should never have to take our word for a result: each writeup states its universe, rules, and costs precisely, and the same backtest pattern — data fetch, signal, loop, stats — is walked through line by line in our code walkthroughs.
Studies
-
Moskowitz, Ooi & Pedersen (2012). Does an instrument's own past 12-month return predict its next month? We replicated it on 10 futures across 4 asset classes.
Result: diversified Sharpe 0.63 (2000–2026) · monthly returns correlate 0.62 with AQR's published factor — the signal replicates; the magnitude decayed post-2009.
Backtests are historical simulations; past results, especially decayed ones, do not predict future returns.