Direct Answer
Factor contribution to risk splits portfolio variance into the part driven by each common factor (market, size, value and so on) and a residual part, using the portfolio's factor exposures and the factor covariance matrix. Large shares show which factors dominate risk; the residual share shows how much risk is stock-specific. It depends on a specific factor model and its estimated covariances, and different models can attribute the same portfolio differently.
Factor Contribution to Risk: Formula, Meaning, Signals, Examples and How to Use It
How Is Factor Contribution to Risk Calculated?
Factor contribution to risk splits portfolio variance into the part driven by each common factor (market, size, value and so on) and a residual part, using the portfolio's factor exposures and the factor covariance matrix.
Formula: FCR_k = b_k x (F b)_k / sigma_p^2; with uncorrelated factors, b_k^2 x sigma_k^2 / sigma_p^2
Inputs:
- Portfolio exposures b_k to each factor
- Factor covariance matrix F
- Idiosyncratic (residual) variance
- Total portfolio variance
Worked example. With exposures of 1.0 and 0.5 to two uncorrelated factors with 15% and 10% volatility and 5% residual volatility, the first factor explains 81.8% of portfolio variance. The inputs are illustrative numbers, not market data.
Reading it. Large shares show which factors dominate risk; the residual share shows how much risk is stock-specific. Unlike risk contribution by asset, this view groups risk by its drivers rather than by holding.
Limits. It depends on a specific factor model and its estimated covariances, and different models can attribute the same portfolio differently. Correlated factors make the split depend on the decomposition used.
What is Factor Contribution to Risk?
Factor contribution to risk splits portfolio variance into the part driven by each common factor (market, size, value and so on) and a residual part, using the portfolio's factor exposures and the factor covariance matrix.
How to use Factor Contribution to Risk
Large shares show which factors dominate risk; the residual share shows how much risk is stock-specific. Unlike risk contribution by asset, this view groups risk by its drivers rather than by holding. A reading alone is never a buy or sell signal; it needs context from price structure, volume, volatility and the broader market regime.
Strengths
- Splitting variance into factor and residual parts shows how much of a portfolio's variability is explained by systematic drivers.
- Shares add to 100 percent, so the share of each factor can be read against the residual directly.
- Reproducible when the formula and data convention are fixed.
Weaknesses and limitations
- A factor's share depends on both exposure and factor volatility, so a modest exposure to a very volatile factor can matter more than a large one to a quiet factor.
- Cross terms between correlated factors are assigned through the covariance form, so a hedge-like factor can show a negative contribution that reads oddly.
- Correlated indicators can create false confidence by repeating the same underlying information.
Editorial and risk note
This page is educational content, not individualized investment advice. It should not imply guaranteed prediction, accuracy or outperformance.
Frequently Asked Questions
Is Factor Contribution to Risk a buy or sell signal?
No. Large shares show which factors dominate risk; the residual share shows how much risk is stock-specific. It depends on a specific factor model and its estimated covariances, and different models can attribute the same portfolio differently. A trade built on it still needs a trigger, an invalidation rule, a position size and realistic execution assumptions.
What is the best setting for Factor Contribution to Risk?
Factor Contribution to Risk is computed from portfolio exposures b_k to each factor; factor covariance matrix F; idiosyncratic (residual) variance, so the choices that matter are the data convention and the window it is read over. Test those across instruments, regimes and out-of-sample periods instead of fitting one historical winner.
Can Factor Contribution to Risk be used by itself?
It depends on a specific factor model and its estimated covariances, and different models can attribute the same portfolio differently. Correlated factors make the split depend on the decomposition used. For that reason it is read next to measures of direction, participation, valuation, liquidity and risk, not alone.
Does Factor Contribution to Risk work on every timeframe?
The calculation (FCR_k = b_k x (F b)_k / sigma_p^2; with uncorrelated factors, b_k^2 x sigma_k^2 / sigma_p^2) can be run on any bar size, but behavior is not portable: a reading on a five-minute chart describes a very different market window from the same reading on a daily chart. Validate the exact timeframe before relying on it.
Why does Factor Contribution to Risk give false signals?
A factor's share depends on both exposure and factor volatility, so a modest exposure to a very volatile factor can matter more than a large one to a quiet factor. Cross terms between correlated factors are assigned through the covariance form, so a hedge-like factor can show a negative contribution that reads oddly. Lag, noise and regime changes add to this, because market participants react to new information after the reading is calculated.
How should Factor Contribution to Risk be backtested?
Rebuild Factor Contribution to Risk from portfolio exposures b_k to each factor using point-in-time data, write the rules before testing, include delisted securities where relevant, model realistic fills and costs, reserve a validation sample and compare with a simpler baseline.