← Research logMicrostructure & Flow
PARTIAL · tool
#

The effective spread can be backed out from the serial covariance of price changes (Roll), giving a transaction-cost estimate without quote data.

$$ \hat s = 2\sqrt{-\mathrm{Cov}(\Delta p_t,\Delta p_{t-1})} $$

Estimate Roll spread per pair, compare to observed quoted spread, use to flag costly markets for the backtester.

Roll vs quoted spreadtracks on liquid names
Breaks when Cov > 0 (trends)yes
PARTIAL EDGE
Useful cost-estimation tool for markets where we lack clean quote history — feeds realistic spreads into the backtester. Not a signal; breaks in trends where serial covariance turns positive (then we fall back to quoted spreads).
Honest backtests need honest costs. When quote data is missing, a covariance-based spread estimate beats assuming zero.

We publish the failures too.

This is one of 100+ documented hypotheses. Browse the full lab notebook, or see the strategies that survived into production.