SPB Git

spb/wp29_uqo

Public

# Data audit

All SHA-256 recomputed with shasum -a 256; all match data/README.md.

file data rows cols first last last value NaN dup dates quarter spacing (days)
vintages/..._2024-01-31.csv 307 2 1947-04-01 2023-10-01 3.3 0 0 90–92
vintages/..._2024-04-30.csv 308 2 1947-04-01 2024-01-01 1.6 0 0 90–92
vintages/..._2024-07-31.csv 309 2 1947-04-01 2024-04-01 2.8 0 0 90–92
vintages/..._2024-10-31.csv 310 2 1947-04-01 2024-07-01 2.8 0 0 90–92
vintages/..._2025-01-31.csv 311 2 1947-04-01 2024-10-01 2.3 0 0 90–92
vintages/..._2025-04-30.csv 312 2 1947-04-01 2025-01-01 -0.3 0 0 90–92
vintages/..._2025-07-31.csv 313 2 1947-04-01 2025-04-01 3.0 0 0 90–92
vintages/..._2025-12-31.csv 314 2 1947-04-01 2025-07-01 4.3 0 0 90–92
current_vintage.csv 318 2 1947-04-01 2026-07-01 3.4 0 0 90–92
first_release.csv 13 3 2023-07-01 2026-07-01 3.4 0 0 —

Units: percent change from preceding quarter, SAAR (percentage points of annualised growth).

Anomalies / notes:

  • COVID quarters 2020Q2 (≈ -28) and 2020Q3 (≈ +35) are extreme outliers in every vintage (genuine data, not errors). Handled in the model by dropping regression observations with dependent quarter in 2020Q2–2020Q4.
  • The 8th origin is 2025-12-31 (not 2025-10-31) because the 2025Q3 advance estimate was delayed by the shutdown; spacing between origins 7 and 8 is 5 months. The h=1 target of origin 8 (2025Q4) was first released in the 2026-02-28 vintage (also delayed).
  • Last value of each vintage equals the first release of that quarter (checked in analysis.py).
  • Revisions within vintages are visible (e.g. 2025Q1: -0.3 → -0.5 → -0.6 → current 0.1).
  • Availability: current_vintage.csv contains post-origin information; it is used only for the revision statistic. first_release.csv is used only as actuals. No input is used in violation of the rules.
Files · main
Menu

Theme