Weekly observations

What the system sees, published before outcomes are known. This page is a permanent, timestamped record of regime classifications and divergence states across 6 regions.

Each observation is a monthly snapshot, cryptographically timestamped at publication. The live API updates daily — classifications may change between monthly observations. Any mid-month regime changes are documented in the following month’s entry. The hash at the bottom of each entry proves what the system reported at publication time.

Version-sensitivity disclosure (2 August 2026): as part of the pre-registered re-testing program, we recomputed the full historical regime record under the current classifier and under the version deployed April–August 2026, from identical input data. Between 18% and 29% of historical months classify differently across the two versions, mostly at the Expansion/Transition boundary. Historical regime sequences are therefore implementation-dependent to a material degree — one more reason this page reports a prospective record rather than a backtest, and why the formal window opens only after re-testing concludes.

Record integrity note (updated 1 August 2026): the initial observation window (opened April 2026) was compromised by mid-window classifier corrections (May 2026) and an infrastructure outage (8 June – 4 July 2026). Those entries remain published for the record but carry no validation weight. The automated publication gate passed its first live run on 1 August 2026; we have deliberately chosen not to open the formal validation window yet. A block-aware statistical re-testing program must conclude first, so that what gets locked has been tested to the standard we intend to be held to. Observations continue to be recorded and timestamped in the interim. When the window opens, parameters lock at that date and the opening entry will say so explicitly. Full methodology at How it works.

Week of 7 April 2026

2026-04-11

Region	Regime	Note
US	Transition	Divergence newly active
AU	Expansion	Growth indicators above expansion threshold. Commodity demand supportive.
DE	Transition	Divergence active. Sovereign yield stress elevated.
GB	Transition	Closest to regime change
JP	Expansion	Changed from Transition — growth indicators crossed expansion threshold despite labour market weakness
CA	Expansion	Leading indicators expanding. Strongest validation z-score of any region.

Divergence

Active in 5 of 6 regions, up from 4 last week. US divergence newly activated. Credit markets remain calm across all regions.

Classification changes

Japan now classified as Expansion, correcting from the Transition classification recorded in the week 14 observation. The week 14 observation was written during a stale data period that affected both GB and JP simultaneously — JP reverted to Expansion on the same April 4 data refresh that corrected GB. JP’s current Expansion classification has been stable since April 4. The conditional destination analysis, which went live during this observation period, shows JP’s current signal conditions as historically resolving to Expansion in the large majority of comparable episodes. This is a post-hoc consistency check, not a prediction confirmed — the destination analysis capability did not exist before the reclassification occurred.

Context

Two new analytical capabilities entered production during this period. The conditional destination analysis draws on 2,308 months of backtested regime history to identify where similar signal conditions have historically resolved. The housing stress transmission layer began its observation period, measuring rate-to-household stress across six signal categories weighted to each region’s mortgage market structure. Initial readings are available in the authenticated platform. Sovereign yield stress remains simultaneously elevated in Germany and Japan — a cross-regional pattern identified in the destination analysis as historically significant. Full signal detail available in the authenticated platform.

Observation hash: b6a209905c2a. RFC 3161 verification token available via API.

Week of 31 March 2026 (corrected)

2026-04-03

Region	Regime	Note
US	Transition	Approaching key boundary — one signal from Stagflation trigger
AU	Expansion	Stable — no elevated boundary pressure
DE	Transition	Low stability — multiple stress channels active
GB	Transition	Closest to regime change — all Stagflation pressure constraints met
JP	Transition	Stable — no elevated boundary pressure
CA	Expansion	Stable — leading indicators above contraction threshold

Divergence

Active in 4 of 6 regions (AU, DE, GB, JP). Energy stress elevated across all regions. Labour market deterioration in DE, GB, and JP independent of the energy channel. Credit markets remain below stress thresholds in all regions. The breadth of simultaneous divergence across 4 regions has not occurred in the backtested validation period (1998–2024).

Classification changes

Corrected observation replacing the original 2026-w14 entry. GB reclassified from Stagflation to Transition after a data refresh resolved stale input signals. All other regions unchanged.

Context

Energy costs remain the dominant cross-regional stress factor. GB is the closest region to a regime change. US is approaching its primary Stagflation trigger. DE mirrors the US directional pattern. The system corrected a stale data issue during this observation period — the directional intelligence (transition pressure) was unaffected and remained consistent throughout. Full correction details available in the authenticated platform.

Observation hash: 64ef15c62030. RFC 3161 verification token available via API.

Week of 31 March 2026 (withdrawn)

2026-04-03

Region	Regime	Note
US	Transition	Approaching stress boundary
AU	Expansion	Stable
DE	Transition	Multiple stress channels active
GB	Stagflation	Labour deterioration
JP	Transition	Labour deterioration
CA	Expansion	Leading indicators declining

Divergence

Active in 4 of 6 regions (AU, DE, GB, JP). Multiple economic stress channels elevated while credit markets remain below stress thresholds. First observation period — no prior baseline.

Classification changes

WITHDRAWN — GB corrected from Stagflation to Transition after data refresh. Corrected observation above (hash 64ef15c62030). Original hash preserved for the record.

Context

Energy costs are the common factor across all six regions. This was the system’s first prospective observation. All classifications were cryptographically timestamped via RFC 3161 before outcomes were known.

Observation hash: 409b747cc946. RFC 3161 verification token available via API.

Observations are published monthly. All classifications are cryptographically timestamped before outcomes are known. Historical observations accumulate on this page as a permanent record of the system’s prospective performance.