Forward-Return Study
Education only · our voice · free public data
Definition
A study of what returns follow a signal or event: for every occurrence, measure the asset's return over the next h periods, then summarize the distribution (mean, median, hit rate, dispersion) across occurrences and horizons.
How to read it
Forward-return studies answer 'when this condition fires, what typically happens next?' The mean forward return is only meaningful alongside its dispersion and sample size - a +1.2% average over 20 events with a 6% standard deviation is statistically indistinguishable from noise. Look at the whole curve of average forward return vs horizon (1d, 5d, 20d, 60d) to see whether the edge builds, peaks, and decays, which tells you the natural holding period. Always separate the MEAN (edge) from the HIT RATE (consistency); positive-skew signals can have <50% hit rate yet positive expectancy.
How practitioners use it
Used as context among multiple indicators — never as a standalone signal to act.
Less common professional uses
Overlapping forward windows (daily-sampled 20-day forward returns) are heavily autocorrelated, so the naive standard error is far too small; use Newey-West/Hansen-Hodrick with lag >= horizon, or sample non-overlapping windows. Event studies are prone to look-ahead in the SIGNAL definition (using revised/restated data, or point-in-time violations); the condition must be computable with only information available at the event timestamp. Conditioning on many candidate signals and reporting the best is multiple testing; the surviving 'edge' needs the same deflation logic as Deflated Sharpe, and a control group (random-entry forward returns) should be shown for comparison. Regime dependence: a positive unconditional forward return can be the average of a strong bull-regime effect and a negative bear-regime effect - stratify by regime before trusting the pooled mean.
Sources & provenance
MacKinlay (1997), 'Event Studies in Economics and Finance', Journal of Economic Literature; Harvey, Liu & Zhu (2016) on multiple-testing hurdles
This page is educational content published by Pachira Aquatica Global LLC. It is not investment advice and not a recommendation.