Data audit
| file | SHA-256 (computed) | matches README | data rows | cols | date range | missing | dup dates | gaps |
|---|---|---|---|---|---|---|---|---|
| UNRATE.csv | 621ff6fadf50910cb7a21ee20ba610c247cb043b29b4e05c4c1a72fadca2a06c | yes | 945 | 2 (observation_date, UNRATE) | 1948-01-01 – 2026-09-01 monthly | 1 (2025-10-01 empty field) | 0 | 0 |
| GDPC1.csv | 1c6c6e9f59e383e20c849dd299175c7a307526c09abdc18f18da9754fc5b3343 | yes | 319 | 2 (observation_date, GDPC1) | 1947-01-01 – 2026-07-01 quarterly | 0 | 0 | 0 |
Units: UNRATE percent (SA); GDPC1 billions chained 2017 $, SAAR. Ranges: UNRATE 2.5–14.8; GDPC1 2172.4–24611.1. All dates are first-of-period, monotonic increasing.
Anomalies: UNRATE 2025-10 is missing (empty field; no October 2025 household survey release). It lies outside the 1959Q4–2019Q4 span needed, so it has no effect; it is recorded in data_issues and removed by the window filter. analysis.py additionally asserts every quarter in 1959Q4–2019Q4 has exactly 3 non-missing months.
Information availability: files contain data after 2019Q4 (and before 1959Q4); these are excluded by the window filter (see out/exclusions.csv). No other violations.