Repository navigation
Record the date the FRED August publication lag closed #136
Description
#127 §7.2. The recent: 1 placement rule in lectures/us_business_cycle_monthly.csv.yml rests on a measurement from 2026-09-07: the live FRED frame ran to 2026-08-01 with only UNRATE and USREC populated for that month. The box asks for the date the lag closed, for the record; the rule stands either way.
Evidence in hand
The Monday canary (run 34092578823, 06:50 UTC on 2026-09-07) validated a frame ending 2026-08-01 with zero revisions in the overlap window, so August's blanks were still there and were accepted. FRED's release calendar puts CPILFESL and INDPRO around the 15th and UMCSENT's final at month end, so the lag should close between 2026-09-15 and 2026-09-26.
To do
Either, from a machine that can reach FRED:
curl -s 'https://fred.stlouisfed.org/graph/fredgraph.csv?id=UMCSENT,CPILFESL,INDPRO' | tail -3
and note the first date on which all three carry a 2026-08-01 value; or read the first Monday canary log after the 15th (the FRED leg prints wrote us_business_cycle_monthly.csv: … (1919-01-01 .. 2026-09-01) once September appears, and the placement rule will then be judging September's blanks, not August's).
Exit criterion
One line on #127 giving the date and which series closed last, and its §7.2 box ticked. If the owner waives this box instead (per the decision item above), close this as not planned and remove it from #132's list.
Activity
mmcky commentedon Sep 7, 2026
Measured 2026-09-07, and the answer is that the lag has not closed, so there is no date to record today.
https://fred.stlouisfed.org/graph/fredgraph.csv?id=UMCSENT,CPILFESL,INDPROends at2026-07-01; there is no 2026-08-01 row at all. Fetched per series as the builder does (one request each, outer-joined):UNRATEandUSRECcarry August,UMCSENT,CPILFESLandINDPROdo not, andM0892AUSM156SNBRended in 1942-06 as its manifest'sended:records. So the live composite runs to 2026-08-01 with exactly two of six series populated in that month — identical to the state therecent: 1rule was written against, and today's canary (run 34092578823) reports the same frame end independently.Next scheduled releases, from each series' FRED page and corroborated against its release calendar:
So the earliest the lag can fully close is 2026-09-25, eighteen days after this close-out. One caveat found in the margin: UMCSENT's 2026-08-28 release delivered no new observation at all — the ALFRED vintages for 08-27 and 08-31 are byte-identical — so its date is the least reliable of the three.
Per the disposition in #133, §7.2 is reworded and ticked rather than waived: the box's substantive claim is confirmed today, and "the date the lag closed" is carried as a named accepted residual on #127. The box's own words are "either way the rule stands", so the date is a record, not a gate — and
recent: 1needs no change.Closing as completed: the check ran and produced a definite answer. If anyone wants the date itself, re-check on or after 2026-09-26.
Generated by Claude Code