Skip to content

KB/signal

NFCI Nowcast

Last verified

Auto-generated. This article is rebuilt from app/signals/config/signal_definitions.json by scripts/build_signals_kb.py. Edit the registry entry and re-run the script — do not edit this file directly.

NFCI Nowcast

What it is

NFCI Nowcast — registry key nfci_nowcast.

Same-week FRESHNESS CARRIER (SIGPRED WS2.1) for the weekly NFCI print, which is days-to-two-weeks stale for most of every publication cycle (week-ending Friday, published the next Wednesday). Each arriving weekly print confirms or denies what the nowcast said the prior session (auto-grading ledger; prereg research/planning/prereg_nfci_nowcast_2026-07-17.md, graded forward as CARRIER_NFCI in nowcast_grading). Not banded / scored / aligned as its own entry; UNLIKE stlfsi4_nowcast it INPUT-FRESHENS the EXISTING nfci leg on BOTH the drawdown-risk lens and the credit-alignment leg (the hy_oas_nowcast precedent – same series / weight / orientation / baseline, an input_source tag, NOT an added leg, so no #96 double-count). Dual-published: the settled nfci print + nfci_report_date are untouched and stay the field of record + grading ground-truth.

Data-pyramid tier

T1 — Signal. This metric sits at the signal layer of the platform’s six-tier data pyramid (T0 raw inputs → T5 actionable read). The tier reflects how far the value is from a raw measurement — not how strongly it is validated. Abstraction and validation are separate axes: a higher tier is not a claim of stronger evidence.

Horizon & validation

No validated skill horizon is on file for this metric — read it as context / a data carrier, not a validated edge. Stamps are added only when a gated research verdict lands.

Source

Source module: computed
Data source: back_fillable

Fetched directly from app/sources/computed.py; see source code for the upstream API call and any provider-specific handling.

How it’s computed

Stale weekly NFCI print plus the frozen OFR-composite bridge times the OFR move since the print’s observation week-end: nfci_nowcast = nfci_stale + 0.01646 * (ofr_fsi_current - ofr_fsi_anchor), where nfci_stale is the row’s stored weekly print, ofr_fsi_current is the row’s OFR FSI composite (T-2 lag – the carrier freshens the up-to-14-day-stale weekly print to T-2, never to today), and ofr_fsi_anchor is the OFR composite as of the print’s observation week-end (the daily_signals row with the LATEST ofr_fsi_obs_date <= nfci_report_date, non-NULL ofr_fsi), resolved by reusing capture._stlfsi4_anchor_ofr. The bridge slope 0.01646 is frozen (capture._NFCI_NOWCAST_BRIDGE): the WS2.1 calibration (2026-07-17, research/studies/ws2_nfci_bridge_calibration.py; NFCI-Friday-aligned weekly deltas, n=792) PRUNED the prereg’s 5-proxy candidate basket (hy_oas / ofr_fsi / stlfsi4_nowcast / vix / move) all the way to the single dominant ofr_fsi proxy – it beats every multi-proxy equal-weight basket in FULL (corr +0.489), STRESS (+0.695), and every era, because the narrower proxies are redundant with the OFR composite (which already aggregates credit / equity-valuation / safe-assets / funding / volatility sub-indices spanning NFCI’s subspace). 0.01646 is the FULL-sample raw dNFCI~dOFR slope; era-stable +0.007..+0.025 across 2011-2026 (SHIP FULL not the 2x-larger thin stress slope 0.0347 – NFCI’s calm weeks carry real OFR-tracking signal, unlike STLFSI4). Abstains (NULL) unless ALL resolve: the stale print, nfci_report_date, ofr_fsi, and the anchor row – no fake fallbacks (a wrong anchor is a confidently wrong carrier). Forward grading: research/planning/prereg_nfci_nowcast_2026-07-17.md (falsifier: forward MAE >= carry OR sign accuracy <= 50% on moved prints over the first ~26 matured weekly prints).

Where it surfaces

Health-score / alignment role

Data carrier — no implication, no health-score contribution.

Persisted for downstream consumers (sparklines, base-rate matcher, calibration substrate) but does not classify into BULLISH / NEUTRAL / BEARISH and does not contribute to the 0-100 health score.

Release cadence

See also