quantmodel — Improve GTT Original corr via vintaged FRED YoY

Objective

Improve GTT Original corr via vintaged FRED YoY.

Notes: ALFRED as-of + either YoY: corr 0.918→0.9999 CAGRΔ +0.03%. Revised conjunction was approximation; AS Redux OR needs vintages. Findings: gtt-original-alfred-asof.md

Approach

Reuses existing report research/findings/gtt-original-alfred-asof.md — tables/metrics preserved verbatim via source_report pointer. Body below summarizes grounded evidence from that report and adjacent artifacts.

Files / code / data changed

Associated commits (inferred from log grep t68):

f81e1a27 research: t68 GTT Original ALFRED asof+either (corr 0.9999) (#66)
f81e1a27 research: t68 GTT Original ALFRED asof+either (corr 0.9999) (#66)
Source report: research/findings/gtt-original-alfred-asof.md
Task-history entry: research/findings/task-history.json#t68

Results

Key table from source report (preserved):

variant	monthly corr	overlap CAGRΔ	MAE
t33 revised FRED + both YoY < 0	0.918	−0.49%	37 bps
revised FRED + either YoY < 0	0.877	−3.48%	46 bps
ALFRED as-of + both	0.918	−0.48%	37 bps
ALFRED as-of + either (shipped)	0.9999	+0.03%	3 bps

Conclusions and metrics are in research/findings/gtt-original-alfred-asof.md; see detail page for preserved tables/metrics.

Conclusions

ALFRED as-of + either YoY: corr 0.918→0.9999 CAGRΔ +0.03%. Revised conjunction was approximation; AS Redux OR needs vintages. Findings: gtt-original-alfred-asof.md

Problems / follow-ups

See source report or PR discussion for follow-ups. No unsupported follow-ups fabricated.