F1-conf-01 — 12-month CPI inflation from two data vintages
Question
12-month growth rate g_t = 100·(P_t/P_{t−12} − 1) of CPIAUCSL (SA) for each month 2001-01 … 2026-09, with P taken from the most recent vintage covering the month, and summaries thereof.
Data and exclusions
cpi_vintage_2016-07-15.csv(834 rows, 1947-01…2016-06) andcpi_vintage_2026-10-28.csv(153 rows, 2014-01…2026-09) — authoritative inputs (n_input_rows).cpi_assembled.csv(957 rows) was read only for auditing. It contains assembly errors: 2014-01…2016-06 taken from the old vintage, 2016-07 missing, and all later values shifted one month later (so it runs to a non-existent 2026-10). It was not used for any computation (data_issues).- The 2026-10-28 vintage has no value for 2025-10 (source reports none). Not imputed, so g for 2025-10 is missing (
missing_growth_months,n_months_missing_growth= 1). (g for 2026-10 would also be affected but lies outside the window.) - The vintages overlap for 2014-01…2016-06 and differ in all 30 months (revisions, max 0.455 index points); the 2026 vintage is used there (
data_issues). - Exclusion ledger (
out/exclusions.csv): 957 monthly levels 1947-01…2026-09 → 309 window months → 308 months with computable g_t (n_analysed).
Method
Combined series: 2026-10-28 vintage for 2014-01…2026-09, 2016-07-15 vintage before. g_t computed by 12-month lag on that series, no interpolation. Summaries are simple arithmetic means over computable months. Deterministic, no randomness (random_seed = null). Code: out/analysis.py.
Results
- Window coverage: first growth month 2001-01 (
first_growth_month), last month with a level 2026-09 (last_month); 308 computable months (n_months_growth), 1 missing (n_months_missing_growth). - Mean 12-month growth over the window: 2.574% (
mean_growth_12m). - Baseline sub-period 2001-01…2013-12: 2.361% (
mean_growth_12m_2001_2013, 156 months,n_growth_2001_2013); 2014-01…2026-09: 2.793% (mean_growth_12m_2014_2026, 152 months,n_growth_2014_2026). - Named months: 2014-06 2.098% (
growth_2014_06); 2016-07 0.868% (growth_2016_07); 2020-05 0.198% (growth_2020_05); 2022-06 8.979% (growth_2022_06); 2026-09 3.717% (growth_2026_09). - Maximum 8.979% in 2022-06 (
max_growth_12m,max_growth_month); minimum −1.959% in 2009-07 (min_growth_12m,min_growth_month). - Feasibility: "partial" (
feasibility,feasibility_reason) because one window month lacks g_t.
Uncertainty
The estimands are deterministic transformations of published index levels; there is no sampling uncertainty model and no confidence intervals are reported. Dispersion of g_t over the window: SD 1.741 pp (sd_growth_12m). The main source of "uncertainty" is data revision: values depend on the vintage (the overlap months differ by up to 0.455 index points between vintages, data_issues).
Limitations and open issues
- g_t for 2014-01…2014-12 mixes a 2026-vintage numerator with a 2016-vintage base (the 2026 extract starts at 2014-01), as the availability rule prescribes; this mixes seasonal-factor vintages.
- 2025-10 g_t is missing; window means cover 308 of 309 months, the 2014–2026 mean covers 152 of 153 months.
- README SHA-256 values refer to raw downloads before reformatting and could not be verified against the provided files.