← All 25 termsQuant methodology
Block bootstrap
QuantA resampling method that preserves short-term autocorrelation in time-series by sampling contiguous blocks rather than individual data points. Used to build honest confidence intervals.
Full definition
A block bootstrap draws resamples by sampling whole contiguous blocks of observations, then concatenating them, instead of drawing individual points independently. The block length is chosen to exceed the longest horizon over which the strategy's returns are autocorrelated (look-ahead, position-holding period, signal-half-life). This preserves the variance and clustering of the original series, which ordinary i.i.d. bootstrap destroys. On this site, block-bootstrap is one of the G1 to G31 evaluation gates. gate G18 uses it to build confidence intervals on Sharpe, drawdown, and turnover without breaking the time-series structure of returns.
Related terms
Each link opens the long-form definition of a closely related term.
See this term in action.
The /methodology page maps every term to the gate or artifact that implements it on this site.