Concept · Stock-level risk (bottom-up)
Factor model
A linear model decomposing an asset's return into common factors (market, sector, …) plus a residual. The foundation of the bottom-up unit.
In depth
RiskModels uses a hierarchical (cascade) factor model: market first (L1), then sector orthogonal to market (L2), then subsector orthogonal to both (L3). Each level adds the incremental exposure, so contributions never double-count shared covariance.
Formula
r_i = α_i + Σ_k β_ik · f_k + ε_i
Compute it with the API
POST /api/decompose
# pip install riskmodels-py
client.decompose("NVDA")In the methodology
The big picture →Referenced by (1)
- ERM3 Cascade-Residual Persistence and the Allocator Skill Signal
Top-decile rank persistence, active-share comparison, and tail-stratified inference across 1,000 top-AUM US mutual funds