What's really inside your fund?
Factor analysis breaks the return of a fund or portfolio down into known sources of return: the stock market, small companies, cheaply valued stocks (value) and momentum. Whatever is left over is the alpha.
Free during the beta, no ads and no subscription. Once you sign in, this page opens the tool.
How it works
It builds on the factors of Eugene Fama and Kenneth French: return differences between groups of stocks, such as small minus big (SMB) or cheap minus expensive (HML). A regression measures how closely your fund's return moves with each factor; those are the loadings.
A loading of 0.3 on value means: if the value factor rises by 1% in a month, your fund is 0.3 percentage points higher on average. The t-stats show how reliable the estimate is; Portfolio Lab also computes them robustly after Newey-West, meaning they hold up against changing volatility and patterns in the residuals.
If a clearly positive alpha remains after all factors, the fund manager did better than simple factor strategies. Often the alpha after costs is negative, and the factors explain almost all of the return.
What a result looks like
Sample dataMade-up sample data, no real prices and no recommendation. Calculated with the same methods as the tool itself.
Inputs in this example
Locked: you can't change the inputs in the preview.
Run your own numbers with a free account Your own inputs, real data and every setting.
Create accountRegression results
t-stats after Newey-West (4 lags). |t| > 2 counts as roughly significant (5% level).
| Factor | Estimate | t-stat | Set in the example |
|---|---|---|---|
| Alpha p.a. | -0.95% | -1.5 | -0.60% |
| Market | 0.94 | 78.3 | 0.95 |
| Size (SMB) | 0.33 | 14.0 | 0.30 |
| Value (HML) | 0.46 | 22.0 | 0.45 |
| Momentum | -0.11 | -7.8 | -0.10 |
The data is made up so that the loadings are known: the regression recovers them up to random noise. Adjusted R² 0.969.
Return attribution
Contribution = loading × avg. factor return p.a. (arithmetic).
| Source of return | Loading | Avg. factor return p.a. | Contribution p.a. |
|---|---|---|---|
| Risk-free rate | – | 2.0% | 2.0% |
| Market | 0.94 | 5.9% | 5.5% |
| Size (SMB) | 0.33 | 3.1% | 1.0% |
| Value (HML) | 0.46 | 3.1% | 1.4% |
| Momentum | -0.11 | 4.6% | -0.5% |
| Alpha (arithmetic) | – | – | -1.0% |
| Total = avg. return p.a. | – | – | 8.5% |
Rolling loadings
Rolling over 36 months: how stable are the factor exposures?
Assumptions: Made-up monthly returns for the fund and the factors, risk-free rate 2% p.a., regression of the excess return with an intercept (OLS).
Frequently asked questions
What is a factor regression?
A regression of a fund's excess return (return minus the risk-free rate) on factor returns. The result is the loading on each factor, the alpha and the R², meaning how much of the volatility the factors explain.
What does alpha mean?
The part of the return that the factors don't explain. An alpha of −1% p.a. means the fund came in one percent a year below what its factor exposure would have led you to expect.
Which factor models are available?
The 3-factor model (market, size, value) and the 5-factor model with profitability and investment, each with momentum and optional bond factors. Factor data is available for the US, Europe, Japan, Asia Pacific and emerging markets.
When is a loading significant?
As a rule of thumb, from a t-stat above 2 or below −2, which is roughly the 5% level. With short time series the estimates are uncertain.
Learn more in the knowledge base
- MethodologyFactor regression, factor allocation and factor statistics: methodology
- ArticleFactor investing: understanding the Fama-French factors
- ArticleBeta and alpha: how closely a fund moves with the market
- ArticleCAPM: the capital asset pricing model and the security market line
- ArticleSharpe's style analysis: what a fund really does
Run it with your own portfolios
Create an account, confirm your email and use this tool with real prices and your own ETFs.
- Price
- Free during the beta
- Data
- Prices, interest rates, inflation and factors from public sources
- Privacy
- Servers in the EU, no ads, statistics without cookies
- Devices
- Computer and phone, everything saved in your account