Skip to content
StockMarketAgent
Tool · Analogs · as of 2026-05-09

Historical analog explorer

What this is warning you about first. the analog result is intentionally withheld until the dataset reaches the minimum history threshold. A premature nearest-neighbour claim would look precise while resting on too little regime history.

Status
Building
No analogs published yet
Coverage
2024-05 to 2026-05
Actual collected history
Snapshots
24 / 36
Minimum before matching
First match
2027-05
Expected threshold month
Closest historical setups

Building 5-year history

Building history
Building 5-year history — first analog match available 2027-05
24 of 36 regime snapshots collected.
Cosine similarity over the standardised macro feature vector requires at least 36 historical snapshots before any past period is a defensible analog. The tool will switch on automatically once that threshold is met.

Forward outcomes (available once history is collected)

Forward-outcome rows will appear once 36 regime snapshots are collected (currently 24).
First analog match available 2027-05. The table is intentionally blank rather than back-filled with synthetic rows — a placeholder forward-return is worse than no forward-return.

No historical analog rows are published yet. The dataset temporal coverage is 2024-05-09/2026-05-09, and the first analog match becomes available in 2027-05.

See the full open data catalog for every dataset behind the research stack.

Frequently asked questions

Why is the analog list blank right now?
The explorer is still building history. It does not publish analog rows until the regime-snapshot ledger reaches the 36-snapshot minimum, because a nearest-neighbour result from a short ledger would be too easy to over-cite.
When does the first match become available?
The current collected coverage runs from 2024-05 to 2026-05, with the first analog match expected in 2027-05 once the required 36 snapshots are present.
Will the explorer use synthetic backfill?
No. The table intentionally stays blank instead of using synthetic rows or retrospective placeholders. The first published analog set will come from the actual snapshot ledger.
How will similarity be computed after launch?
The ready-state explorer will compare the current macro feature vector against prior snapshots using standardised breadth, valuation, credit, rotation, and regime-score inputs. Until the threshold is met, the page publishes the method but not any match claims.
Every tool here is a thin lens on the same evidence the per-ticker reports use. Inputs, scoring rules, and exclusions are documented in the methodology. Read the full methodology.