Concept · Stock-level risk (bottom-up)
Return attribution
Splitting a stock's *realized return* into contributions from each factor. Distinct from variance: a name can be variance-dominated by market yet return-driven by residual.
In depth
Splits a realized return (not variance) into per-factor contributions. Over multiple periods the contributions compound, so RiskModels uses a geometric bridge to keep them additive.
Compute it with the API
GET /api/returns-decomposition
# pip install riskmodels-py
client.get_returns_decomposition("NVDA")In the methodology
Multi-period attribution: the geometric bridge →Referenced by (0)
No working paper references this concept yet.