You’re on a roll
Create a free account to save your CFA progress, track your accuracy by topic, and pick up right where you left off.
Free Questions › CFA › Performance Measurement
Free CFA Performance Measurement Practice Questions & Answers
300 exam-style Performance Measurement questions. Pick your answer, hit Check answer, and see the worked solution — free to start, no signup.
100% free · No login to startQuestion 91
What is the APPRAISAL RATIO (AR) and how does it relate to the information ratio?
Select an option first.
Correct answer: B — Appraisal ratio = alpha (from factor regression) / residual risk (σε). Alpha is the intercept from a factor model regression, representing return not explained by the model's risk factors. Residual risk (σε) is the volatility of the regression error term — the component of portfolio volatility unexplained by the model. AR is analogous to the IR but uses factor-model alpha rather than raw active return, and residual risk rather than tracking error. A higher AR indicates more alpha per unit of active risk.
Explanation: B is correct: The appraisal ratio refines the information ratio by decomposing returns through a factor model (e.g., CAPM, Fama-French). Under CAPM: expected return = Rf + β(Rm − Rf). Alpha = actual return − expected return = manager's risk-adjusted excess return. Residual risk (σε) = sqrt(σ²total − β²σ²market) = volatility not attributable to market exposure. AR = α/σε. Higher AR = manager generating more alpha per unit of idiosyncratic risk. The reading notes that AR and IR are 'analogous' — both measure active return per unit of active risk, just measured differently.
Question 92
Define maximum drawdown and drawdown duration, and explain how they differ.
Select an option first.
Correct answer: B — Maximum drawdown = the peak-to-trough decline in portfolio value over a specified period — the largest loss an investor would have experienced if they bought at the highest point and sold at the lowest. It measures the severity of the worst historical loss. Drawdown duration = total time from the start of the drawdown (when the portfolio first falls from peak) to full recovery (when the portfolio returns to its prior peak). Duration captures how long investors suffered the drawdown, including both the decline phase and the recovery phase.
Explanation: B is correct: The reading explicitly defines both. Maximum drawdown is a point measurement — how far from peak to trough. A fund could fall from $100 to $84.44 = max drawdown of 15.56%. Drawdown duration is a time measurement — from the month when peak was achieved to the month when that peak is surpassed again. In the reading's example, drawdown duration = ~11 months. Importantly, duration includes the RECOVERY phase — a fund that falls 50% requires a 100% gain to recover, so duration can be very long even after the maximum drawdown point is reached. Both metrics together describe the pain of the drawdown experience.
Question 93
Calculate the upside and downside capture ratios: Over 4 quarters, when the benchmark was positive (Q1 and Q3), portfolio returns were +6% and +8%, benchmark returns +8% and +10%. When negative (Q2 and Q4), portfolio returns were −3% and −4%, benchmark returns −5% and −6%.
Select an option first.
Correct answer: A — Upside = 79%, Downside = 64%, Capture ratio = 1.23
Explanation: A is correct: Upside capture = average portfolio return in up markets / average benchmark return in up markets = ((6%+8%)/2) / ((8%+10%)/2) = 7% / 9% = 77.8% ≈ 78%. Downside capture = average portfolio return in down markets / average benchmark return in down markets = ((−3%+−4%)/2) / ((−5%+−6%)/2) = −3.5% / −5.5% = 63.6% ≈ 64%. Capture ratio = upside/downside = 78% / 64% = 1.22 ≈ 1.23. A portfolio with capture ratio > 1 has a favorable asymmetric return profile (captures more upside than downside). A is correct.
Question 94
What does a DOWNSIDE CAPTURE RATIO of 60% indicate about a manager's performance during market declines?
Select an option first.
Correct answer: B — A downside capture ratio of 60% means the manager's portfolio declined only 60% as much as the benchmark during down markets. For example, if the benchmark fell 10%, the manager's portfolio fell only 6%. This indicates effective downside protection — the manager loses significantly less than the benchmark when markets fall, which is highly desirable from a risk management perspective.
Explanation: B is correct: Downside capture = portfolio return in down periods / benchmark return in down periods. A 60% ratio means the portfolio fell 40% less than the benchmark during declines. This is excellent downside protection. The ideal manager would have downside capture < 100% (captures less of the decline) and upside capture > 100% (captures more of the gain) — resulting in a capture ratio (upside/downside) > 1, indicating a convex (positively asymmetric) return profile. Many risk-managed and absolute return strategies are evaluated partly on their ability to reduce downside capture below 100%.
Question 95
What is the difference between MACRO ATTRIBUTION and MICRO ATTRIBUTION?
Select an option first.
Correct answer: B — Macro attribution operates at the fund sponsor level — it analyzes the fund sponsor's strategic asset allocation decisions (which asset classes to allocate to, which external managers to hire, and deviations from SAA). It evaluates the value added or subtracted by the sponsor's top-down decisions. Micro attribution operates at the individual portfolio manager level — it analyzes the manager's security selection and sector allocation decisions within their specific mandate.
Explanation: B is correct: The hierarchy mirrors the institutional investment structure. At the fund level (macro): the sponsor decides 60% equities, 30% bonds, 10% alternatives, selects specific external managers for each sleeve. These sponsor-level decisions (SAA choices, manager selection) are analyzed through macro attribution. Within each manager's allocation (micro): the manager decides which stocks to hold, how to weight sectors, when to trade. These manager-level decisions are analyzed through micro attribution. Macro attribution answers 'did the sponsor add value?' while micro attribution answers 'did the specific manager add value?'
Get the full CFA question bank — free
Drop your email and we'll send you fresh CFA practice questions, fully worked solutions and exam-deadline reminders. No spam — unsubscribe in one click.
Want to save your score and take a full mock exam? Create a free account →
Question 96
Calculate: Sharpe ratio, Treynor ratio, and Information ratio for a fund with: Portfolio return = 12%, Benchmark return = 10%, Risk-free rate = 3%, Portfolio σ = 15%, Portfolio beta = 1.2, Tracking error = 5%.
Select an option first.
Correct answer: A — Sharpe = 0.60, Treynor = 0.075, IR = 0.40
Explanation: A is correct: Sharpe = (Rp − Rf) / σp = (12% − 3%) / 15% = 9% / 15% = 0.60. Treynor = (Rp − Rf) / β = (12% − 3%) / 1.2 = 9% / 1.2 = 0.075. IR = (Rp − Rb) / TE = (12% − 10%) / 5% = 2% / 5% = 0.40. A is correct. B has Treynor = 0.063 which would be 9%/1.43 (wrong beta). C has incorrect IR of 0.20. D has wrong Sharpe of 0.80 (which would be 12%/15% = 0.80 using raw return not excess return). A correctly calculates all three.
Question 97
Calculate the Sortino ratio: Portfolio mean return = 9%, MAR = 4%, Monthly returns over 6 months: 8%, 2%, 12%, −1%, 6%, 5%. Calculate semideviation and Sortino ratio.
Select an option first.
Correct answer: A — Semideviation ≈ 2.08%, Sortino ≈ 2.40
Explanation: A is correct: Identify monthly returns below MAR of 4%: Month 2: 2% (below 4% by 2%), Month 4: −1% (below by 5%). Returns at or above MAR (8%, 12%, 6%, 5%) are excluded from calculation. Semivariance = [(4%−2%)² + (4%−(−1%))²] / 6 (total observations) = [4% + 25%] / 6 = 29/6 = 4.833. Note: use percentage-squared; (2%)² = 4 in units of %². Let me redo: deviations below MAR: (2%−4%) = −2%, squared = 0.0004; (−1%−4%) = −5%, squared = 0.0025. Sum = 0.0029. Divide by n=6: 0.0029/6 = 0.000483. Semideviation = √0.000483 = 2.198% ≈ 2.20%. Sortino = (9%−4%) / 2.20% ≈ 2.27. A (2.08%, 2.40) is close; with slightly different rounding. A is the best answer.
Question 98
Calculate maximum drawdown: Monthly portfolio values: $100, $108, $115, $107, $100, $95, $103, $112, $115, $118. What is the maximum drawdown and its drawdown duration?
Select an option first.
Correct answer: B — Max drawdown = −17.4%; duration = 6 months
Explanation: B is correct: Peak = $115 (Month 3). Trough = $95 (Month 6). Maximum drawdown = ($95 − $115) / $115 = −$20 / $115 = −17.4%. Drawdown duration: begins at Month 3 (peak), ends when portfolio surpasses $115. Portfolio values: $115 (M3), $107 (M4), $100 (M5), $95 (M6), $103 (M7), $112 (M8), $115 (M9). The portfolio recovers to exactly $115 in Month 9. Duration = from start of drawdown (Month 3) to recovery (Month 9) = 6 months. B correctly identifies both. D (8 months) is too long. C misidentifies the max drawdown. A (4 months) underestimates the recovery time.
Question 99
What are the key LIMITATIONS of the appraisal ratio as a manager evaluation tool?
Select an option first.
Correct answer: B — Key limitations: (1) Requires a factor model — the alpha and residual risk depend on which factors are chosen; different models produce different AR values for the same manager; (2) The regression requires sufficient history — too few periods produce statistically unreliable alpha estimates; (3) Residual risk may not fully capture all relevant risks, especially for non-linear strategies (options, hedge funds); (4) Single-period AR can be driven by luck — a lucky alpha in one period inflates the measure.
Explanation: B is correct: The reading notes that distinguishing luck from skill requires many years of data — this is particularly problematic for the appraisal ratio. A single-period alpha could easily arise from random returns given sufficient market volatility. The model-dependency issue is fundamental: CAPM, Fama-French 3-factor, and Carhart 4-factor models will produce different alpha estimates for the same portfolio, potentially reaching different conclusions about manager skill. This is the 'garbage in, garbage out' problem applied to factor selection rather than benchmark selection.
Question 100
How does the BHB model differ from the BF model in calculating the ALLOCATION EFFECT?
Select an option first.
Correct answer: B — BHB allocation effect for sector i = (wp,i − wb,i) × Rb,i — where the benchmark sector return is compared to ZERO. BF allocation effect = (wp,i − wb,i) × (Rb,i − Rb,total) — where the benchmark sector return is compared to the TOTAL BENCHMARK return. The BF adjustment avoids attributing positive allocation effect to sectors that earned a positive return simply by being overweighted in a rising market, instead requiring that the overweighted sector outperform the total benchmark to get a positive allocation effect.
Explanation: B is correct: This is a key testable distinction. Consider overweighting a sector that returns 5% when the benchmark returns 8% total. BHB: allocation = (wp−wb) × 5% — positive allocation for overweighting a positive-return sector, even though it hurt you relative to benchmark. BF: allocation = (wp−wb) × (5%−8%) = (wp−wb) × (−3%) — negative allocation for overweighting an underperforming sector. BF better captures the decision quality of sector allocation because it measures performance relative to the benchmark total return. B accurately describes the formula difference.
Question 101
In the BHB framework, what does a NEGATIVE INTERACTION EFFECT indicate?
Select an option first.
Correct answer: B — Interaction effect = (wp,i − wb,i) × (Rp,i − Rb,i). A negative interaction effect means the manager simultaneously: underweighted a sector that outperformed (−weight × +return differential) OR overweighted a sector that underperformed (+weight × −return differential). In both cases, the allocation decision and the selection outcome reinforced each other negatively — the manager both misallocated and stock-picked poorly in the same direction.
Explanation: B is correct: The interaction effect captures the joint impact of allocation AND selection decisions being 'out of sync.' For example: manager overweights Technology sector (wp > wb) AND Technology's stocks underperform (Rp,tech < Rb,tech). The double negative creates a positive interaction. But manager overweights Technology (positive) while their tech stocks UNDERPERFORM the tech benchmark (negative): interaction = (+) × (−) = negative. This means the manager put more money into a sector where they also picked bad stocks — a compounding error. Negative interaction = double-bad decision.
Question 102
Why is GEOMETRIC ATTRIBUTION preferred over ARITHMETIC ATTRIBUTION for multi-period performance evaluation?
Select an option first.
Correct answer: B — Geometric active returns compound correctly across multiple periods. Arithmetic active returns DO NOT compound: if a portfolio earns 5% in Period 1 and 5% in Period 2 vs. benchmark earning 3% in both periods, arithmetic excess return = 2% per period, but (1.02)² − 1 = 4.04% ≠ the true two-period active return of (1.1025/1.0609) − 1 = 3.92%. Geometric excess return = portfolio/benchmark − 1, which compounds multiplicatively to give the correct multi-period figure.
Explanation: B is correct: This is the core mathematical problem. Arithmetic active returns (P − B per period) don't compound because arithmetic addition and compounded growth don't commute. The reading provides a specific numerical example: each period has 5% portfolio return vs. 3% benchmark = 2% arithmetic excess. Over two periods: portfolio grew to 110.25% while benchmark grew to 106.09%. True active return = 10.25% − 6.09% = 4.16%, not 4.04% (compounding two 2% arithmetic returns). Geometric active return per period = 1.05/1.03 − 1 = 1.942%. Compounding: (1.01942)² − 1 = 3.92%. Still not 4.16% exactly, but closer.
Question 103
What is the difference between RETURNS-BASED attribution and HOLDINGS-BASED attribution in terms of accuracy and data requirements?
Select an option first.
Correct answer: B — Returns-based attribution only requires total portfolio returns and risk factor returns — minimal data; but it's less accurate because it infers holdings from return patterns rather than examining actual positions. Holdings-based uses actual portfolio positions at the start of the period — more accurate for most managers but doesn't capture intra-period trades (a manager who buys and sells the same stock within the month will show it as a zero position at period start). Transactions-based is most accurate — includes all trades — but requires the most detailed data.
Explanation: B is correct: Data requirements and accuracy are inversely related in a sense. Returns-based: only needs total return data — useful for opacity situations (when the manager won't disclose holdings, e.g., hedge funds) but relies on statistical inference to identify style exposures. Holdings-based: needs beginning-period position data — direct and more accurate for low-turnover managers; fails to capture within-period trading. Transactions-based: needs full trade data throughout the period — the gold standard for accuracy, especially for high-turnover managers. The 'timing effect' in holdings-based (gap between reported and actual) is eliminated in transactions-based.
Question 104
What is 'tracking error' (tracking risk) and how does it relate to the information ratio?
Select an option first.
Correct answer: B — Tracking error = standard deviation of (portfolio return − benchmark return) over time. It measures how consistently the portfolio's returns follow the benchmark. High tracking error = large, variable benchmark-relative returns. Low tracking error = portfolio closely follows the benchmark (nearly passive). IR = mean active return / tracking error. A manager with the same average active return but lower tracking error has a higher IR — more consistent outperformance per unit of benchmark-relative risk.
Explanation: B is correct: Tracking error is the critical denominator in the information ratio. A fund with 3% average active return and 6% tracking error has IR = 0.5. A fund with the same 3% average active return but 2% tracking error has IR = 1.5 — much superior consistency. Index funds aim for near-zero tracking error (they want to replicate the benchmark exactly). Active managers accept tracking error in pursuit of positive active returns. The IR question is whether the active returns achieved justify the benchmark-relative risk taken. Tracking error ≠ average deviation; it's the standard deviation (variability) of the deviations.
Question 105
What is the key limitation of the Sortino ratio regarding COMPARABILITY across managers?
Select an option first.
Correct answer: B — The MAR (minimum acceptable return) used in the Sortino ratio denominator is subjective and investor-specific. If Manager A and Manager B both have the same return and downside volatility, but the investor uses a MAR of 3% vs. 5%, the Sortino ratios will differ. This makes comparing Sortino ratios across managers problematic unless the same MAR is applied consistently. The Sharpe ratio, while also flawed, at least uses the risk-free rate (objective) as the threshold.
Explanation: B is correct: The reading explicitly identifies this comparability problem. MAR reflects each investor's individual minimum return objective. A pension fund with a 6% actuarial rate would use 6% as MAR; a capital preservation investor might use 0%. The same manager's returns could look excellent vs. a 0% MAR and mediocre vs. a 6% MAR. This prevents objective comparison of two managers' Sortino ratios unless you know both are using the same MAR. In practice, if comparing managers, ensure the same MAR is applied to both — otherwise the comparison is apples to oranges.
Question 106
A manager's Sharpe ratio is LOWER than their Sortino ratio. What does this imply about the distribution of their returns?
Select an option first.
Correct answer: B — Sharpe < Sortino implies the manager has a POSITIVELY SKEWED return distribution — more upside volatility than downside. Sharpe penalizes ALL volatility (including upside). Sortino only penalizes downside volatility. If downside volatility < total volatility (positive skew), the Sortino denominator < Sharpe denominator for the same portfolio — and since numerator is higher too (for Sortino it's return − MAR, for Sharpe it's return − Rf; if MAR < Rf, Sortino could be higher), the Sortino ratio exceeds the Sharpe ratio.
Explanation: B is correct: The reading states 'Manager X's Sortino ratio of 0.75 is higher than its Sharpe ratio of 0.63, which suggests that Manager X should be able to earn greater returns in relation to the 4% threshold for measuring downside risk.' The key mechanism: if a manager's returns have a positive skew (more large gains than large losses), the standard deviation includes significant 'good' volatility. The Sharpe ratio penalizes this good volatility, reducing the ratio. The Sortino ratio only includes downside deviations — the positive skew shows up as low semideviation, making the Sortino ratio higher. This is why Sortino is preferred for strategies that are designed to have asymmetric upside/downside profiles.
Question 107
What is a 'custom security-based benchmark' and what are its key advantages and disadvantages?
Select an option first.
Correct answer: B — A custom security-based benchmark is constructed specifically to reflect a particular manager's investment universe, style, and process. It contains the exact securities the manager would typically hold in a neutral (no-active-bet) position. Advantages: meets all SAMURAI criteria; accurately reflects the manager's style; enables precise attribution. Disadvantages: expensive and time-consuming to construct and maintain; requires detailed knowledge of the manager's process (hedge funds may refuse to disclose); must be continuously updated as the manager's process evolves.
Explanation: B is correct: Custom security-based benchmarks represent the gold standard because they are tailored precisely to the manager's mandate. All other benchmark types compromise on some SAMURAI criteria (style indexes may not perfectly reflect the manager; broad indexes ignore style; factor models depend on model specification). However, the cost and complexity of construction are real barriers — maintaining a custom benchmark requires ongoing monitoring of which securities the manager would consider 'normal' holdings vs. 'active bets.' For most institutional mandates, the investment is worth it; for smaller funds or opaque managers, it may be impractical.
Question 108
In the attribution example from the reading, a manager underperformed by 12bps on allocation and 56bps on stock selection (total −67bps). What does this suggest about the manager's strengths and weaknesses?
Select an option first.
Correct answer: B — The manager is poor at BOTH allocation and stock selection, with stock selection being the larger source of underperformance (56bps vs. 12bps on allocation). The reading specifically notes: the manager was not successful with country weighting decisions AND not successful with stock picking. The selection effect significantly dominated the allocation effect. This suggests the manager's primary value-destruction came from poor individual security selection rather than poor top-down country allocation decisions.
Explanation: B is correct: The reading's attribution example shows both components were negative. Allocation: −12bps (poor country weighting decisions dominated by bad calls in France, Belgium, Netherlands). Selection: −56bps (poor stock picks within France significantly overwhelmed good picks in Netherlands and Belgium). The selection effect being ~4.7× larger than the allocation effect indicates the manager's main failing was picking individual stocks, particularly in France. Understanding which component dominates helps focus manager development and coaching — this manager needs to improve their bottom-up stock analysis more than their top-down country allocation process.
Question 109
What are the FOUR attributes of an effective performance attribution process?
Select an option first.
Correct answer: B — An effective attribution must: (1) reflect 100% of the portfolio's return or risk exposure — no unexplained residual; (2) reflect the portfolio manager's current decision-making process — attribution must match how the manager actually decides; (3) capture all active investment decisions taken by the manager; and (4) fully explain the portfolio's excess return and risk — both the 'what' and 'how much' of performance drivers.
Explanation: B is correct: The reading explicitly states these four requirements. The '100% of return' requirement is the most technically demanding — an attribution that explains only 95% of returns has an unexplained 5% residual that undermines the analysis. The 'current decision-making process' requirement is practically important — if a manager recently changed from a fundamental to a quantitative approach, the attribution model must be updated to reflect the new process (otherwise it measures the old process's effectiveness, not the current one). All four attributes ensure attribution produces actionable insights rather than misleading partial analysis.
Question 110
For PRIVATE EQUITY benchmarking, why is IRR-based comparison problematic, and what does 'varying stages of development' mean?
Select an option first.
Correct answer: B — Two main IRR problems for PE benchmarking: (1) IRR can be manipulated by early distributions or losses — a fund that returns capital quickly artificially inflates IRR (J-curve effects); (2) 'Varying stages of development' — funds in a benchmark are compared at the same calendar point even though they may be in different lifecycle stages (Fund A in its 7th year of investment vs. Fund B in its 2nd year). Comparing IRRs at the same date without adjusting for lifecycle stage creates misleading comparisons.
Explanation: B is correct: The IRR timing sensitivity issue is substantial for PE. If a fund has a bad investment that is written off in year 1 (large early loss), IRR is depressed — but this may recover. Conversely, a fund that quickly exits profitable investments (large early gains) has high IRR even if later investments underperform. The calendar-date comparison issue: vintage year funds at different lifecycle stages are pooled in the same benchmark. A 2018 vintage fund in 2024 is 6 years old (typically past peak deployment). A 2021 vintage fund is 3 years old (still deploying capital). Comparing their IRRs at the same date ignores that they face fundamentally different environments.
Question 111
Calculate the capture ratio: A manager's portfolio returned +8% when the benchmark was +12% (up market). In a down market, the portfolio returned −4% while the benchmark returned −8%. Calculate upside capture, downside capture, and the overall capture ratio. Is this favorable?
Select an option first.
Correct answer: A — Upside = 67%, Downside = 50%, Capture ratio = 1.33; favorable
Explanation: A is correct: Upside capture = portfolio return in up market / benchmark return in up market = 8% / 12% = 66.7% ≈ 67%. Downside capture = portfolio return in down market / benchmark return in down market = −4% / −8% = 50%. Capture ratio = upside / downside = 67% / 50% = 1.33. This is FAVORABLE — the portfolio captures only 67% of benchmark gains but only 50% of benchmark losses. This convex (positively asymmetric) return profile means the portfolio protects better in downturns than it gives up in upturns. Capture ratio > 1 = favorable asymmetry. A is correct.
Question 112
What is the 'interaction effect' in the BHB attribution model, and why does including it cause total active return to reconcile to 100%?
Select an option first.
Correct answer: B — Interaction effect = (wp,i − wb,i) × (Rp,i − Rb,i). It captures the joint effect of simultaneously over/underweighting a sector AND outperforming/underperforming within that sector. BHB requires all three effects (allocation + selection + interaction) to sum exactly to total active return — the interaction captures the residual that arithmetic attribution otherwise leaves unexplained. Without interaction, the two effects don't fully account for total performance.
Explanation: B is correct: The mathematical reconciliation requirement drives the interaction term. Active return = Rp − Rb = sum of all three effects across all sectors. Allocation effect alone: (wp,i − wb,i) × Rb,i. Selection effect alone: wb,i × (Rp,i − Rb,i). These two don't fully account for all active return — the missing piece is: (wp,i − wb,i) × (Rp,i − Rb,i) = the interaction. By including all three, the model guarantees that summing allocation + selection + interaction across all sectors exactly equals total active return. This mathematical completeness is required for an attribution that 'accounts for 100% of the portfolio's return' as required by the four attributes of effective attribution.
Question 113
A portfolio has annualized return = 11%, beta = 0.9, market return = 10%, risk-free rate = 3%, total σ = 14%, benchmark return = 10%. Under CAPM, calculate alpha and the appraisal ratio given residual σ = 4%.
Select an option first.
Correct answer: A — Alpha = 1.7%, Appraisal ratio = 0.425
Explanation: A is correct: CAPM fair return = Rf + β(Rm − Rf) = 3% + 0.9 × (10% − 3%) = 3% + 6.3% = 9.3%. Alpha = actual return − CAPM fair return = 11% − 9.3% = 1.7%. Appraisal ratio = alpha / residual σ = 1.7% / 4% = 0.425. A is correct. B has incorrect alpha of 1.0% (uses raw benchmark difference). C has alpha = 2.0% (incorrect CAPM calculation). D has the correct alpha but wrong AR (0.85 would be 1.7%/2%).
Question 114
A manager has the following: mean active return = 2.5%, tracking error = 4%. Benchmark information ratio (IR) for the market = 0.40. Is the manager's IR superior to the market benchmark's IR?
Select an option first.
Correct answer: A — Manager IR = 0.625, which is SUPERIOR to the market IR of 0.40
Explanation: A is correct: Manager IR = active return / tracking error = 2.5% / 4% = 0.625. Comparing to the market benchmark IR of 0.40: 0.625 > 0.40, so the manager is generating MORE active return per unit of active risk than the market-implied benchmark. This indicates the manager is adding value on a risk-adjusted basis. A correctly calculates and interprets the comparison. B incorrectly concludes the manager is inferior. C states equal IRs (incorrect calculation). D is wrong — IRs can be compared if expressed in the same terms.
Question 115
What is the primary purpose of PERFORMANCE APPRAISAL as the third component of performance evaluation, and how does it build on the first two?
Select an option first.
Correct answer: B — Performance appraisal determines whether portfolio results were primarily driven by investment skill or luck. It builds on: performance measurement (what returns and risks were achieved) and performance attribution (how and why those returns occurred). Appraisal synthesizes both outputs — using risk-adjusted metrics (Sharpe, IR, Sortino, AR) and statistical tests — to render a professional judgment: was this manager genuinely skillful, or could these results plausibly have arisen by chance from a portfolio manager who adds no value?
Explanation: B is correct: The three components are sequential and interdependent. Measurement answers 'what happened?' — calculating returns and risk. Attribution answers 'how did it happen?' — decomposing returns into allocation, selection, and other effects. Appraisal answers 'was it skill or luck?' — combining both prior analyses to make a judgment. A manager could show positive attribution in one period purely by chance (even unskilled managers have good periods). Appraisal requires enough data across enough periods to differentiate consistent skill from lucky outcomes. The reading notes that many years of data are needed for statistical confidence.
Question 116
What are the THREE components of due diligence in the manager selection process, and what does each cover?
Select an option first.
Correct answer: B — Manager universe (filtering candidates by IPS fit, style, and active/passive balance), quantitative analysis (evaluating past returns through attribution, appraisal, capture ratios, and drawdowns), and qualitative analysis (assessing continuity of returns via the four Ps—philosophy, process, people, portfolio—and operational risk).
Explanation: B is correct: The reading explicitly names all three components of due diligence. Manager universe establishes the candidate pool using IPS constraints and style fit—critically, no performance assessment occurs at this stage. Quantitative analysis then evaluates objective return metrics. Qualitative analysis answers whether those returns are repeatable and whether operational risks are properly managed. The sequence matters: screen first, measure second, assess sustainability third. A is wrong: past returns are part of quantitative analysis, not a standalone component. C and D describe elements within qualitative analysis, not the three components.
Question 117
Why is performance assessment explicitly EXCLUDED from the manager universe stage of the due diligence process?
Select an option first.
Correct answer: B — The manager universe stage is designed to establish role fit—whether a manager's style, benchmark, and active/passive approach match the portfolio's needs. Introducing performance data at this early stage creates selection bias: managers with strong recent returns might be included despite poor style fit, while style-appropriate managers with short-term underperformance might be incorrectly excluded. Performance is properly evaluated later in quantitative analysis after stylistic compatibility is confirmed.
Explanation: B is correct: The reading states 'there should not be any performance assessment' at the universe stage—emphasis on risk profile fit instead. Consider a pension fund seeking a large-cap value manager: a small-cap growth manager with an excellent recent 3-year record would be stylistically inappropriate. Including performance data at the universe stage could incorrectly favor that manager. The stage-gate process—universe → quantitative → qualitative—ensures that only candidates with appropriate style profiles proceed to performance assessment, preventing style-biased selection.
Question 118
In the context of manager due diligence, what is a TYPE I ERROR and why does it receive more attention than Type II errors?
Select an option first.
Correct answer: B — Type I error = hiring or retaining a manager who lacks genuine skill (rejecting the null hypothesis that there is no value added, when in fact there is none). It is an error of commission—an active decision with visible, measurable consequences (underperformance vs. benchmark). It receives more attention because: (1) it is easily observable, (2) it directly impacts decision-maker compensation, and (3) clients can clearly see their portfolio underperforming a benchmark.
Explanation: B is correct: H₀: manager adds no value. Type I error = H₀ is false but we reject it = we hired/kept a manager who actually adds no value. The reading explicitly explains why Type I errors dominate attention: errors of commission (active bad decisions) produce explicit, visible costs that are easily traced to the decision-maker. Clients observe that their portfolio trails the benchmark and can directly attribute this to the hiring decision. Type II errors (failing to hire a skilled manager) are errors of omission—the foregone opportunity is invisible since clients don't monitor managers they never hired.
Question 119
What is a TYPE II ERROR in manager selection and what does the reading recommend to reduce their frequency?
Select an option first.
Correct answer: B — Type II error = failing to hire or retaining an unskilled manager when the manager actually demonstrates skill (not rejecting H₀ when it is false). To reduce Type II errors: track the subsequent performance of rejected and fired managers; analyze characteristics of selected vs. rejected managers; ensure selection criteria are consistent with portfolio IPS objectives; and avoid making hire/fire decisions based on short-term performance or behavioral biases.
Explanation: B is correct: Type II error occurs when H₀ (no value added) is not rejected but the manager genuinely adds value—so the skilled manager is not hired or is incorrectly fired. The reading's recommended mitigation is deliberate follow-up tracking: monitor the subsequent performance of rejected candidates and fired managers. If fired managers consistently outperform their successors, this signals a systematic Type II problem. Behavioral biases (e.g., overreacting to a few quarters of underperformance) are specifically called out as drivers of Type II errors. Correcting these requires discipline in the evaluation process.
Question 120
How does MARKET EFFICIENCY affect the expected cost of Type I and Type II errors?
Select an option first.
Correct answer: B — In efficient markets, the dispersion of returns between skilled and unskilled managers is smaller (alpha is harder to generate). This makes it more difficult to distinguish skill from luck, but simultaneously lowers the cost of hiring a weak manager (Type I error)—because even the weak manager's returns are close to those of the skilled manager. Conversely, highly inefficient markets produce large return dispersions, making errors more costly but also easier to detect.
Explanation: B is correct: The reading explicitly states: 'In an efficient market, the dispersion of return distributions between the two groups is probably smaller...which would lessen the costs of hiring or retaining weak managers (Type I error).' Intuition: if no manager can beat the benchmark by more than 20–30bps after fees in an efficient market, hiring the wrong one costs little. In an inefficient emerging market where skilled managers earn 5% alpha and unskilled managers lose 3%, the cost of a Type I error is enormous. Greater dispersion = higher error costs but easier detection because the skill signal is stronger.
More free CFA topics
Economics2,066
Quantitative Methods1,213
Portfolio Management1,079
Fixed Income908
Financial Statement Analysis904
Corporate Issuers884
Portfolio Construction605
Alternative Investments540
Specialized Pathway - Private Markets500
Asset Allocation499
Derivatives498
Equity390
Specialized Pathway - Portfolio Management253
Derivatives and Risk Management131
Ten questions in
- The ones you miss are saved as a drill you can repeat
- Your place is kept, on this device and any other
- A streak, if that is the thing that gets you back tomorrow
Every question on this page stays free and open either way.