Independent research on systematic trading. Methods, sources and limitations included.
Latest publication 12 October 2026
Only with your consent. Unsubscribe at any time.
Latest report
This report examines how statistical inference changes when monthly backtests use standard errors adjusted for serial correlation, and how much a signal depends on when its inputs became observable.
All reports are listed below.
Earlier reports
A US equity/cash rule's annualized residual falls from 2.37% to 0.22% under six-factor controls and a 25 bp cost scenario.
A historical US-market backtest examines information timing, rule selection, revised data, and assumed trading costs.
A descriptive comparison of VIX-implied volatility and subsequent S&P 500 realized risk across daily 30-day windows and selected episodes.
Annual US returns and a separate monthly yield proxy
Only with your consent. Unsubscribe at any time.
Brent and WTI regional spot prices, oil benchmark and contract terms, and European power-market time units and annual turnover.
Investment-grade and high-yield return covariance, private-credit annual returns and borrower metrics, with their measurement limits.
Why similar Sharpe ratios and equal volatility can conceal very different loss tails and drawdowns.
A walk-forward study of portfolio risk forecasts, their errors in market stress and the sources of the gap.
Evidence on signal decay before and after publication, compared by discovery cohort, data type and strategy capacity.
Published stock-return predictors compared across reconstructed backtest, forward and post-publication windows. These are not verified live trading results.
How annual management fees reduce the growth that remains with the investor over a long holding period.
Historical strategy-index returns across interest-rate regimes, separating market conditions from strategy behaviour.
A plain-language guide to the kinds of AI used in quantitative trading and how they fit into an investment system.
How AI changes operating models and competition, and what separates defensible alpha from a marketing narrative.
Research updates
Receive new research by email, with a copy of our latest report. Only with your consent. Unsubscribe at any time.
Published for information, not personal investment advice. Research methodology
Next: Start here