Skip to main content
All CollectionsOnboarding
Understanding Factor Analysis Output
Understanding Factor Analysis Output
M
Written by Mei Chung
Updated over 5 months ago

Rev. Date July 29, 2024

Risk factors are discrete, describable sources of common or systematic risk and return across a diverse set of investments. Venn offers multiple ways to view factor analysis, and it’s important to utilize all of them to better understand the story of your investments or portfolios.

The Factor Exposures block shows the sensitivity of the portfolio or investment to the factors within the Two Sigma Factor Lens. The Factor Contributions to Risk block decomposes which of those factors are driving the risk of the portfolio or investment. The Factor Contributions to Return block confirms if those risks are being compensated, which is arguably where the rubber meets the road.

What do factor exposures tell me?

The factor exposure chart shows a portfolio or investment’s sensitivity to the factors within the Two Sigma Factor Lens. Factor exposures provide useful insights that can help with portfolio management including, but not limited to, the following:

  1. Confirm that the exposures are in agreement with expectations or stated objective (e.g., a Value equity fund having positive sensitivity to the Value factor)

  2. Confirm how diversified a portfolio is (e.g., an Alternatives portfolio with a high Equity factor exposure may indicate that it’s actually not as diversifying as intended to be relative to a traditional Equity portfolio)

  3. Identify any unwanted exposures or gaps between target allocations and what’s actually been realized (e.g. you are looking for a Core Equities manager but the manager actually has material style tilts)

At the portfolio level, you can see further details on the drivers of the factor exposures both on investment and strategy levels by clicking onto the factor exposures. This is helpful to confirm that the factor exposure drivers are consistent with expectations or to identify what is driving any unintended exposures.

How are factor exposures calculated and what do they mean?

Venn uses a two-step regression process to calculate the exposures. First, the full factor lens is run through a Lasso regression to select a subset of the factors that are most relevant.[1] Then an ordinary least squares (OLS) regression is used on the filtered set of factors to compute the final Betas and t-stats that are shown in Venn. Factors with t-stats whose absolute value is smaller than 1.96 or with less than 1% contribution to risk will be considered statistically insignificant and have faded bars in the factor analysis charts.

A positive (negative) exposure (ß) to a factor indicates that the portfolio/investment’s returns are moving in the same (opposite) direction as the factor, and the size of the exposure provides insights into the magnitude of the movement (i.e., the larger the Beta, the more sensitive the portfolio/investment is to the factor’s movements).

For more details on how to interpret exposures to individual factors, please refer to the Factor Interpretation Guide here.

Factor Exposure Example

Demo Portfolio Factor Exposures

Source: Venn by Two Sigma. Benchmark used is iShares MSCI ACWI ETF. Period from 12/20/2011 - 7/22/2024.

The chart above shows the average annualized factor exposures of Venn’s Demo Portfolio over the total Analysis Period. From this chart we can see that the Demo Portfolio has positive exposures to the Equity and Interest Rates factors, which is common for a traditional portfolio.[2] As shown below, the Equity exposure is driven by the Equity strategy within the portfolio, which is consistent with expectations.

Demo Portfolio Equity Factor Exposure Contribution by Strategies

Source: Venn by Two Sigma. Period from 12/20/2011 - 7/22/2024.

The portfolio also has a positive beta to the Local Equity factor, indicating that the portfolio has a home country bias as it is overweight Domestic Equities (the U.S. Equities in this example) relative to World Equities, which is measured by the broader Equity factor. In terms of style factors, the portfolio has a slight positive exposure to Value, which is driven by the two Value funds in the portfolio, confirming that the funds are consistent with their stated objectives.

Demo Portfolio Value Factor Exposure Contribution by Investments

Source: Venn by Two Sigma. Period from 12/20/2011 - 7/22/2024.

Active Factor Exposure Example

Venn subscribers often analyze performance of long-only portfolios on a relative to benchmark basis. This can be done in Venn by toggling on the “Relative to Benchmark” field at the top of the Analysis page or within Configuration. Active factor exposures are then calculated from the excess return stream of the portfolio or investment and will show positioning of the portfolio/investment relative to the selected benchmark.

Equity Sub-Portfolio Active Factor Exposures

Source: Venn by Two Sigma. Benchmark used is iShares MSCI ACWI ETF. Period from 12/20/2011 - 7/22/2024.

The above chart shows the active factor exposures of the Equity sub-portfolio of the Demo Portfolio with respect to the iShares MSCI ACWI ETF. From this chart, we can see that the Equity Portfolio has an overweight to Local Equity (U.S. equities), a negative style tilt to Quality, and positive style tilts into Value, and Small Cap relative to the benchmark.

From a magnitude perspective, the active exposures are roughly flat or moderate, indicating that the portfolio exposures aren’t deviating significantly from the benchmark.

What are the factor risk contributions and why are they important?

A burrito is a commonly loved food that has all the major food groups covered. However from a nutritional perspective, a burrito has high levels of sodium and fat. Similarly, a portfolio may be diverse from an asset class perspective, but the “nutrients” may not be as diversified when considering the common risk factors driving that portfolio.

Similarly, a portfolio may look diversified from a factor exposure perspective, but not from a risk perspective as factors have varying levels of volatility.

Factor contribution to risk provides insights into the breakdown of portfolio risk by its systematic (factors) and idiosyncratic (residual) components. Monitoring factor contributions to risk can help allocators identify the common risk and return drivers across asset classes and security types to achieve diversification more effectively.

What is residual risk?

The residual risk shows how much of the portfolio’s risk is the idiosyncratic risk that isn’t explained by the Two Sigma Factor Lens. Residual risk and returns can serve as helpful metrics for justifying the fees of active or alternative managers and evaluating their performance.

Factor Risk Contribution Example

Demo Portfolio Factor Exposures

Source: Venn by Two Sigma. Benchmark used is iShares MSCI ACWI ETF. Period from 12/20/2011 - 7/22/2024.

Demo Portfolio Factor Contribution to Risk

Source: Venn by Two Sigma. Benchmark used is iShares MSCI ACWI ETF. Period from 12/20/2011 - 7/22/2024.

Revisiting the Demo Portfolio, we can see that the total portfolio has material exposures to both the Equity and Interest Rates factors. However, about 82% of the risk in the portfolio is driven by the Equity factor, indicating that the portfolio isn’t as diversified from a risk perspective.

Equity Sub-Portfolio Active Factor Contribution to Risk

Source: Venn by Two Sigma. Benchmark used is iShares MSCI ACWI ETF. Period from 12/20/2011 - 7/22/2024.

Looking at the active factor risk contributions of the Equity sub-portfolio, we can see a material residual contribution to the excess returns, which can be considered a positive sign for active managers. Similarly, we would expect to see high residual contributions to risks for diversifying strategies such as alternatives or private investments.

What do factor return contributions tell me?

Now that you analyzed the factor exposures of your portfolio/investment and how much of the risk they define, you may want to know if those risks are being compensated. The factor contributions to return measures the amount of return attributable to each risk factor and residual. In conjunction with the factor exposures, return contributions will inform the performance impact of the portfolio or investment’s positioning.

Factor Return Contribution Example

Demo Portfolio Factor Exposures

Source: Venn by Two Sigma. Benchmark used is iShares MSCI ACWI ETF. Period from 12/20/2011 - 7/22/2024.

Demo Portfolio Factor Contribution to Risk

Source: Venn by Two Sigma. Benchmark used is iShares MSCI ACWI ETF. Period from 12/20/2011 - 11/23/2023.

Demo Portfolio Factor Contribution to Return

Source: Venn by Two Sigma. Benchmark used is iShares MSCI ACWI ETF. Period from 12/20/2011 - 7/22/2024.

Looking at the Demo Portfolio as an example, the high risk contribution from the Equity factor was compensated as the Equity factor defines the majority of the total portfolio returns over the analysis period and the factor performed positively over the analysis period. The residual, however, detracted from performance. The positive exposures to Local Equity also paid off as the factor was up over the analysis period. However, the positive exposures to the Foreign Currency and Small Cap risk factors generated losses as they were down over the analysis period.

Active Factor Return Contribution Example

Equity Sub-Portfolio Active Factor Contribution to Return

Source: Venn by Two Sigma. Benchmark used is iShares MSCI ACWI ETF. Period from 12/20/2011 - 7/22/2024.

Looking at the factor contributions to excess returns for the Equity sub-portfolio as another example, the positive tilt into Small Cap relative to the iShares MSCI ACWI ETF hurt the sub-portfolio’s excess returns as the factor was down over the analysis period. Meanwhile, the overweight to Local Equity helped performance relative to the benchmark as the factor was up over the analysis period. The underweight to Foreign Currency also contributed positively to excess returns, as the factor was down over the analysis period, thus the underweight resulted in fewer losses to the sub-portfolio compared to the benchmark.

[1] Factors with 0 beta coefficients after the Lasso regression are considered insignificant and are excluded from the analysis. Excluded factors are footnoted at the bottom of each factor analysis chart.

[2] A common traditional portfolio is comprised of roughly 60% Equities and 40% Bonds.

This document highlights certain aspects of this feature. As an overview, it does not discuss all material facts or assumptions. Please see Important Disclosure and Disclaimer Information.

Did this answer your question?