Factor Intelligence is in beta — the compute methodology is still being validated. All factor responses carry the SECAPI-Maturity: beta response header; do not yet treat returned values as production-grade.
What Factor Intelligence Covers
SEC API Factor Intelligence provides a structured factor model for launch-focused U.S. market, style, sector, and industry workflows. The API supports factor catalog retrieval, single-factor history, multi-factor sparklines, historical and intraday returns, extreme move detection, correlation analysis, per-stock exposure loadings, regime-conditioned performance, portfolio decomposition, and consolidated dashboard views.
Every response includes requestId and traceparent for downstream traceability.
For the full factor construction methodology, including purification architecture, academic references, and institutional comparisons, see Factor Model Methodology. For source posture and freshness contracts, see Factor Provenance and Factor Freshness.
Factor catalog
- launch-focused factors are organized into market, style, sector, and industry categories
- each factor entry includes methodology description, source status, proxy or construction metadata, orthogonalization metadata, and category assignment
- the catalog is the canonical reference for factor identifiers used across all other factor endpoints
- launch factors target a
2015-01-01 public history floor; shorter-history or beta factors are labeled as such and should not be used for unsupported coverage claims
Categories
Factor returns
- historical daily factor return series with z-scores and volatility-scaled variants
- supports filtering by factor category, individual factor identifiers, and date range
- used for backtesting, trend analysis, and factor-timing research
- use
response_mode=compact when forwarding results into an agent, chart, or table workflow
Query parameters
Single-factor history and sparklines
- single-factor daily history with 1D through MAX summary windows, bounded by row-level coverage and freshness metadata
- compact multi-factor sparkline series for sector, index, model, and dashboard UI views
- designed for token-efficient chart/table rendering where the caller needs returns plus window summaries, not full attribution payloads
Extreme moves and pairs
- extreme factor moves ranked by z-score, return, direction, and threshold
- extreme factor pairs ranked by unusual spread divergence
Factor correlations
- factor-to-factor and factor-to-security correlation matrices
- supports windowed correlation with configurable lookback
- useful for strategy diversification analysis and risk decomposition
Query parameters
Factor exposures
- per-stock, portfolio, or watchlist factor loadings with model metadata and provenance
- exposures reflect the sensitivity of a security’s returns to each factor
- supports single-ticker and batch queries
Query parameters
Factor intraday
- intraday factor return snapshots derived from the current benchmark proxy inputs when the market-data plane is current
- intended for dashboards and real-time factor monitoring during market hours
- callers should inspect freshness metadata before treating an intraday snapshot as current
Factor dashboard
- consolidated one-call view combining intraday returns, regime context, rotation signals, and available factor state
- designed to power full-screen factor dashboards in a single request
- factor rankings conditioned on the current macro regime
- blends the active macro backdrop (expansion, contraction, stress, recovery) with current factor state
- useful for regime-aware factor rotation and allocation decisions
Query parameters
Portfolio factor analysis
- decompose an arbitrary portfolio into factor exposures, return attribution, and hedge suggestions
- accepts a list of holdings with weights in the request body
- returns explained return, residual alpha, and per-factor contribution
Portfolio attribution and hedge
- dedicated attribution endpoint for explained return, alpha, and compact contribution rows
- dedicated hedge endpoint for constrained hedge candidates and residual factor exposure
- both endpoints accept the same normalized holdings shape used by portfolio analysis
Model factor analysis
- ad hoc Model Builder or saved-model holdings analysis without first persisting a model portfolio
- optional attribution, hedge, and per-position sections in one compact response
- useful for
/models and model detail views that need the same factor intelligence as portfolio workflows
Endpoint summary
Errors and partial results
- Treat non-2xx responses as API errors and log the response body for debugging
- Log
requestId and traceparent so SEC API support can trace the request
- if provenance or freshness metadata is present, return it unchanged so trust is not lost in the handoff