Concept · Stock-level risk (bottom-up)
Variance decomposition
Splitting a stock's variance into shares attributable to each factor and the residual. Sums to 1.0.
In depth
The shares are orthogonal, so they sum to 1.0 within tolerance — no double-counting. This is the single-stock version of the portfolio decomposition.
Formula
1.0 ≈ l3_market_er + l3_sector_er + l3_subsector_er + l3_residual_er
Compute it with the API
GET /api/l3-decomposition
# pip install riskmodels-py
client.get_metrics("NVDA", as_dataframe=True)[["l3_market_er","l3_sector_er","l3_subsector_er","l3_residual_er"]]In the methodology
Explained risk: variance decomposition →Referenced by (1)
- The Persistence of Stock-Selection Residuals
A point-in-time, holdings-based decomposition of mutual fund performance — the stock-selection residual persists out of sample; style timing and sector timing, measured the same way, do not.