Free lesson · Point-in-time data
First prints and revisions can reverse a signal
Open interactive lessonPractice calculationsExplore labs
Start with the idea
A revised series answers what is believed now about the past; a first-print series answers what was reported then.
Symbols, units & horizon
- x^(v₁),x^(v₂): values for the same observation period in earlier and later vintages
- Δx: revision in the original series units
- v₁,v₂: ordered publication/version dates, not two economic observation periods
When and why to use this
Audit macro nowcasts and release strategies for look-ahead through revised values.
A revised series answers what is believed now about the past; a first-print series answers what was reported then.
Preserve a row for each observation period and vintage. A later revision should not rewrite the historical model’s input. Comparison of real-time and latest-vintage signals is a useful diagnostic of how much apparent predictability came from revised information.
A data-provider vintage date may have day-level granularity. Intraday studies additionally need release-time evidence. If that evidence is unavailable, delay trading to a conservative later time and label the timing approximation.
First prints and revisions can reverse a signal
- Hold the economic observation period fixed.
- Read its value from each archived vintage.
- Subtract earlier from later; do not call the result economic growth between periods.
First-reported monthly growth is −0.2 percentage points and the revision is +0.1. Revision size is .1−(−.2)=+.3 percentage points; the sign-based first-print signal differs.
Apply it in a strategy
- Freeze inputs at the stated decision time and record their units.
- Audit macro nowcasts and release strategies for look-ahead through revised values.
- Recompute the example, then change the material assumption and explain the difference.
Research deliverable
First prints and revisions can reverse a signal: produce the worked calculation, a timestamped input record and a written decision addressing this limitation: A vintage comparison alone does not establish precise release timing or explain the economic cause of a revision.
These are synthetic mechanics examples, not historical performance or paper replications. Module evidence and research boundaries record the 12 September 2026 review.
Python implementation
Self-contained teaching example. Python 3.10+; dependencies and input conventions are shown in the code and notation. Run in your own Python environment.
# Python 3.10+; standard library and NumPy only.
# Synthetic teaching inputs; conventions and units are defined in the notation above.
def revision(first,later):
return later-first
assert abs(revision(-.2,.1)-.3)<1e-12
print(revision(-.2,.1))Continue learning
Data: Availability, Revisions & Reproducible Research — all lessons- Four clocks for one observation
- An as-of join that never selects the future
- First prints and revisions can reverse a signal
- Reconcile splits and cash distributions
- Universe membership and disappearing assets
- Coverage, missingness and stale values
- Purging labels that cross a test boundary
- Reproducible snapshots and discrepancy ledgers
Quantitative finance and development glossary · Python resources and libraries · Research sources and limitations