Skip to content

CFA Level II Portfolio Management Practice Questions

Official outline
CFA Institute 2026 Level II Topic Outlines (learning outcomes)
Questions on this page
8
Updated
Nov 3, 2026

  1. Question 1, medium

    A portfolio has a total VaR of USD 9.0 million, split among three risk components: equity USD 5.5 million, credit USD 2.4 million and interest rate. What is the component VaR of interest rate?

    • A.USD 2.2 million
    • B.USD 1.1 million
    • C.USD 7.9 million
    • D.USD 16.9 million
    Show answer and explanationHide answer

    Correct answer: B. USD 1.1 million

    Component VaRs add up to total VaR. Interest rate component = 9.0 - 5.5 - 2.4 = USD 1.1 million. Component VaR attributes diversified portfolio risk, so it differs from stand-alone VaR figures that would not sum to the total.

    • A. Doubles the correct answer with no basis.
    • B. Correct: total less the other two components.
    • C. That is the sum of equity and credit components.
    • D. That sums all three figures and ignores that the total is given.

  2. Question 2, medium

    An analyst builds a factor model in which the factors are standardized company attributes such as book-to-price, earnings yield and size. Each stock's sensitivity is its attribute value, and factor returns are estimated by cross-sectional regression each period. What type of model is this?

    • A.Macroeconomic factor model
    • B.Statistical factor model
    • C.Single-index market model
    • D.Fundamental factor model
    Show answer and explanationHide answer

    Correct answer: D. Fundamental factor model

    In a fundamental factor model, factors are company attributes and the sensitivities are those attributes, with factor returns estimated each period. Macroeconomic models use surprises in variables such as inflation with sensitivities from time-series regression, and statistical models extract factors from return data.

    • A. Macroeconomic models use economy-wide surprises, not company attributes.
    • B. Statistical models extract factors from the return covariance, not from attributes.
    • C. The single-index model has one market factor.
    • D. Correct: attribute-based factors with cross-sectional estimation.

  3. Question 3, medium

    A portfolio has sensitivities of 1.2 to the market factor, 0.5 to a size factor and -0.4 to a value factor. Factor risk premiums are 5.0% for the market, 2.0% for size and 3.0% for value, and the risk-free rate is 3.0%. What is the expected return under the multifactor model?

    • A.7.8%
    • B.8.8%
    • C.5.8%
    • D.11.2%
    Show answer and explanationHide answer

    Correct answer: B. 8.8%

    Expected return = 3.0 + 1.2 x 5.0 + 0.5 x 2.0 + (-0.4) x 3.0 = 3.0 + 6.0 + 1.0 - 1.2 = 8.8%. The negative value sensitivity reduces expected return because the value premium is positive.

    • A. Omits the size factor contribution.
    • B. Correct: risk-free rate plus the sum of sensitivity times premium.
    • C. Omits the risk-free rate.
    • D. Treats the value loading as positive.

  4. Question 4, hard

    A manager runs a Monte Carlo simulation of a multi-asset portfolio. Correlations are estimated from a calm five-year period and returns are drawn from normal distributions. Which result is most likely?

    • A.Simulated tail losses are understated, because correlations rise and returns are fat-tailed in crises
    • B.Simulated tail losses are understated for single-asset holdings but overstated for diversified ones
    • C.Simulated tail losses are overstated, because calm periods contain extreme volatility
    • D.Simulated tail losses are accurate, because the number of trials removes input error
    Show answer and explanationHide answer

    Correct answer: A. Simulated tail losses are understated, because correlations rise and returns are fat-tailed in crises

    Monte Carlo output is only as good as its inputs. Calm-period correlations and normal draws miss that diversification weakens in stress and that extreme moves are more common than normality implies. More trials reduce sampling error, not input error.

    • A. Correct: input assumptions drawn from calm markets understate crisis risk.
    • B. Diversified portfolios are especially exposed when correlations rise.
    • C. Calm periods contain little extreme volatility.
    • D. More trials do not correct poor model inputs.

  5. Question 5, easy

    A back-test assigns each firm's annual earnings to the fiscal year-end date, although the figures are not published until three months later. Which bias does this introduce?

    • A.Look-ahead bias
    • B.Data-snooping bias
    • C.Survivorship bias
    • D.Time-period bias
    Show answer and explanationHide answer

    Correct answer: A. Look-ahead bias

    Look-ahead bias arises when a simulated strategy uses information that was not available at the decision date. Using figures before publication lets the model trade on future knowledge, which inflates results.

    • A. Correct: data used before it was public.
    • B. Data-snooping comes from repeatedly testing many specifications on the same data.
    • C. Survivorship bias comes from excluding failed or delisted firms.
    • D. Time-period bias comes from results that depend on the chosen sample window.

  6. Question 6, medium

    Using a Gordon growth approach, an analyst sets expected equity return equal to dividend yield plus long-run growth. The dividend yield is 2.0%, long-run growth is 4.3% and the risk-free rate is 2.8%. What is the implied equity risk premium?

    • A.4.3%
    • B.9.1%
    • C.6.3%
    • D.3.5%
    Show answer and explanationHide answer

    Correct answer: D. 3.5%

    Expected return = 2.0% + 4.3% = 6.3%. Equity risk premium = expected return - risk-free rate = 6.3% - 2.8% = 3.5%. Reporting 6.3% confuses total expected return with the premium.

    • A. Reports only the growth term.
    • B. Adds the risk-free rate instead of subtracting it.
    • C. Reports expected return without subtracting the risk-free rate.
    • D. Correct: 6.3% expected return less the 2.8% risk-free rate.

  7. Question 7, easy

    A researcher back-tests a small-cap strategy using only the stocks in today's small-cap index, including their full price history. Firms that went bankrupt or were delisted during the test period are absent. How does this affect the back-test?

    • A.It introduces no bias, because delisted stocks have no future returns
    • B.It introduces survivorship bias that tends to understate returns
    • C.It introduces survivorship bias that tends to overstate returns
    • D.It introduces look-ahead bias that tends to understate risk but leave returns unchanged
    Show answer and explanationHide answer

    Correct answer: C. It introduces survivorship bias that tends to overstate returns

    Excluding failed firms leaves out poor performers, so the sample looks better than the investable universe at the time. Back-tested returns are overstated and risk is understated. Using the constituents as of each historical date avoids it.

    • A. Delisted stocks had returns while they traded; dropping them biases the sample.
    • B. Gets the direction wrong; removing losers raises results.
    • C. Correct: only survivors remain, which biases performance upward.
    • D. The problem is the missing failed firms, not information timing, and returns are affected too.

  8. Question 8, hard

    Under a single-factor arbitrage pricing model, two well-diversified portfolios, Kestrel and Larch, each have a factor sensitivity of 1.0. Kestrel has an expected return of 8.0% and Larch has an expected return of 6.5%. Which statement is correct?

    • A.No arbitrage exists, because the portfolios may differ in firm-specific risk
    • B.An arbitrage exists: buy both portfolios and finance with the risk-free asset
    • C.An arbitrage exists: buy Larch and sell Kestrel short
    • D.An arbitrage exists: buy Kestrel and sell Larch short
    Show answer and explanationHide answer

    Correct answer: D. An arbitrage exists: buy Kestrel and sell Larch short

    Well-diversified portfolios have negligible firm-specific risk, so equal factor exposure should give equal expected returns. Buying the higher-return portfolio and shorting the lower-return one has zero net factor exposure and a positive expected profit.

    • A. Diversification eliminates specific risk, so it cannot explain the difference.
    • B. Buying both does not remove factor risk and is not an arbitrage.
    • C. Takes the position on the wrong side, locking in a loss.
    • D. Correct: long the higher-return portfolio and short the lower one, with zero net factor risk.

Questions are drafted with help from AI and published only after a person reviews and approves them. How these questions are made

More Portfolio Management questions

Practice Portfolio ManagementCustomize CFA Level II practice

Keep going

The same subject on other exams

More CFA Level II topics

Already working in the field?

Learn about Parsonex