FRM Exam Part II · Portfolio Construction

Factor-Based Portfolio Construction and Smart Beta

Updated 11 October 2026 · Fact-checked

Factor-based portfolio construction builds portfolios to hold chosen risk factor exposures (value, momentum, low volatility) instead of picking individual securities. Smart beta does this with rules-based, transparent, cheap indices. To solve questions, find the factor betas, multiply by factor returns, add them up, and compare with the benchmark.

Understand Factor-Based Portfolio Construction and Smart Beta

A factor is a driver of returns shared across many securities, such as market, value, size, momentum, quality or low volatility. Factors explain why stocks move together and why some earn higher long-run returns. Each stock's return can be split into a part explained by factors and a leftover idiosyncratic part.

Factor portfolio construction starts from the exposure you want. You pick the factor, rank securities on a characteristic (for example book-to-market for value), and form a portfolio tilted toward it. There are two main forms. A long-only factor portfolio overweights high-ranked stocks against a benchmark. A long-short portfolio buys the top group and sells the bottom group. It is roughly market-neutral, so it isolates the pure factor premium. It needs shorting, leverage and financing, which add cost and risk.

Smart beta products are rules-based indices or funds that tilt toward factors or use alternative weights (equal weight, fundamental weight, minimum variance) instead of market-cap weights. They sit between passive and active management. Like passive, they are rules-based, transparent and low cost. Like active, they deviate from the cap-weighted market. Unlike traditional active management, the deviations come from a systematic rule, not manager judgement.

A factor risk model gives portfolio exposure to each factor and splits total risk into factor risk and specific risk. Portfolio factor exposure is the weighted average of the holdings' exposures. Diversification works across factors too. Factors with low or negative correlation (value and momentum are often cited) can reduce total risk when combined. But correlations can rise in stress, so factor diversification is not guaranteed. Factor returns are also cyclical and can underperform for years, and crowding and costs can erode the premium.

Key formulas to remember

Single-factor/multifactor return: Rp = α + β1·F1 + β2·F2 + … + βk·Fk + ε; Fk are factor returns, βk are exposures, ε is idiosyncratic return.
Portfolio factor exposure: βp,k = Σ wi · βi,k; Weights times each holding's exposure to factor k. Use active weights (portfolio minus benchmark) for active exposure.
Long-short factor return: RLS = R(long leg) − R(short leg); Approximately market-neutral. Excludes financing and borrowing costs unless stated.
Systematic variance: σp² = β′ Σf β + σε²; Σf is the factor covariance matrix. Two factors: β1²σ1² + β2²σ2² + 2β1β2σ1σ2ρ.
Share of risk from factors: Factor share = systematic variance ÷ total variance; Equals R² of the factor regression when estimated that way.

How to solve Factor-Based Portfolio Construction and Smart Beta questions

Use this order for most factor and smart beta questions.

1Identify what is asked: exposure, expected return, risk split, or choice of strategy.
2List the factors, the betas (or characteristics) and the factor returns or volatilities given.
3Compute portfolio exposure as the weighted sum of holding betas, using active weights if a benchmark is involved.
4Multiply each beta by its factor return and add alpha if given to get the factor-explained return.
5For risk, apply β′Σβ, including the correlation term, then add specific variance and take the square root for volatility.
6For long-short, use long minus short and check the net market exposure and costs.
7Interpret: is the tilt intended, is risk concentrated in one factor, and does diversification hold in stress?
8Check the answer options for the trap: signs, squared terms, or variance versus volatility.

Quickest way: Beta times factor, then add

When to use it: Return attribution or exposure questions with few factors.

Write each factor's beta and return in a column.
Multiply pairs and sum; do not forget the sign of short or negative betas.
For risk, work in variance; take the square root only at the end.
Eliminate options that ignore the correlation term or use weights that do not sum to the stated total.

Common mistakes in Factor-Based Portfolio Construction and Smart Beta

Treating smart beta as the same as active management
Both deviate from market-cap weights.
Fix: Smart beta follows transparent systematic rules at low cost; active relies on discretionary judgement.
Adding factor volatilities instead of variances
Rushing through the risk formula.
Fix: Add variances with the covariance term, then take the square root.
Assuming a long-short factor portfolio has no risk
Market-neutral is read as risk-free.
Fix: It still has factor, crowding, leverage, short-squeeze and financing risk.
Assuming factor diversification always holds
Low average correlations are remembered.
Fix: Correlations can rise in stress and factor premiums can be negative for long periods.
Using total weights instead of active weights for tracking exposure
Benchmark is forgotten.
Fix: Active exposure equals portfolio exposure minus benchmark exposure.

Worked examples

Example 1

A portfolio has betas of 1.0 to market, 0.4 to value and −0.2 to momentum. Factor returns for the period are market 6%, value 3% and momentum 5%. Alpha is 0.5%. What is the expected return from the model?

Show the solution

Market: 1.0 × 6% = 6.0%
Value: 0.4 × 3% = 1.2%
Momentum: −0.2 × 5% = −1.0%
Sum factors: 6.0 + 1.2 − 1.0 = 6.2%
Add alpha: 6.2% + 0.5% = 6.7%

Answer: 6.7%

Example 2

A portfolio has exposures of 0.5 to factor A and 0.5 to factor B. Factor A volatility is 10%, factor B volatility is 20%, and correlation is 0. Specific volatility is 3%. What is total portfolio volatility?

Show the solution

Factor A variance: 0.5² × 0.10² = 0.25 × 0.01 = 0.0025
Factor B variance: 0.5² × 0.20² = 0.25 × 0.04 = 0.0100
Correlation term is zero
Specific variance: 0.03² = 0.0009
Total variance: 0.0025 + 0.0100 + 0.0009 = 0.0134
Volatility: √0.0134 ≈ 0.1158

Answer: About 11.6%

Exam tips

Read whether the question wants variance or volatility, and whether weights are active.
Know the difference between long-only tilts and long-short factor portfolios, including costs and leverage.
Expect conceptual items on smart beta versus passive and active, and on factor crowding and cyclicality.
Check the sign of each beta before multiplying.
Remember factor diversification can weaken in stress.

Practice questions from Portfolio Construction

Factor-Based Portfolio Construction and Smart Beta in other exams

The same ground in other exams, if you are preparing for more than one or want another angle on it.

Factor-Based Portfolio Construction and Smart Beta: frequently asked questions

What is the difference between smart beta and active management?

Smart beta follows a transparent, rules-based approach to tilt toward factors or alternative weights, usually at lower cost. Active management relies on manager judgement to select securities. Smart beta is often seen as a middle ground between passive and active.

How do you construct a long-short factor portfolio?

Rank securities on the factor characteristic, buy the top group and short the bottom group. Weights are set to make the portfolio roughly market-neutral. Its return is the long leg minus the short leg, before financing and trading costs.

What does a factor risk model tell you?

It shows the portfolio's exposure to each factor and splits risk into factor and specific parts. This helps you see whether risk is concentrated in one factor. It also supports attribution and benchmark-relative risk analysis.

Do factor premiums always pay off?

No. Factors can underperform for long periods, and their returns can fall when strategies become crowded. Costs and stress-period correlation changes also matter.