Overview
Performance Measurement is 5-10% of the CFA Level 3 exam and has three learning modules. Portfolio Performance Evaluation separates measurement, attribution and appraisal: it covers return and risk attribution, equity and fixed-income attribution output, liability-based and asset-based benchmarks and their quality tests, and appraisal measures such as the Sortino ratio, appraisal ratio, capture ratios and drawdown. Investment Manager Selection covers due diligence, Type I and Type II errors, style analysis, vehicle and contract choice, and performance-based fees. The third module is an overview of the Global Investment Performance Standards (GIPS).
The topic closes the loop on everything else in the curriculum. Asset Allocation and Portfolio Construction make decisions; this topic determines which decision-maker, the asset owner or the manager, was responsible for each part of the result and whether that result reflects skill.
How it's tested
Questions present a performance report, an attribution table, a benchmark description or a manager's fee terms and ask what they reveal. Item sets favor computing a ratio or reading attribution effects; constructed-response items may ask you to evaluate a benchmark against quality criteria, recommend a manager or vehicle, or judge whether a firm's practice complies with GIPS.
-
Interpret attribution output
An equity or fixed-income attribution table is given and you identify which decision (allocation, selection, duration or curve positioning) drove relative performance.
-
Choose and compute an appraisal measure
Returns, downside deviation, residual risk or drawdown data are given and you compute the measure that answers the stated concern.
-
Judge a GIPS or selection practice
A firm's composite policy or a consultant's manager decision is described and you decide whether it complies or which error it risks.
A manager's attribution shows a large positive allocation effect and a negative selection effect, yet the client chose the sector weights through a policy benchmark. The question asks how much value the manager added. Because the asset owner set the allocation, only the selection effect reflects the manager's decisions, so the manager actually detracted value despite the positive total excess return.
Learning modules 3
Official learning modules from the CFA Institute outline, in outline order.
3.1 Portfolio Performance EvaluationMeasurement, attribution, appraisal, benchmarks and risk-adjusted measures.
- What it means
- This module defines performance measurement (what the return was), attribution (why it differed from the benchmark) and appraisal (whether it reflects skill), then covers return and risk attribution, returns-, holdings- and transactions-based methods, fixed-income attribution, liability-based and asset-based benchmarks, benchmark quality, and appraisal measures.
- What to understand
- Macro attribution measures the asset owner's decisions such as the strategic allocation, while micro attribution measures the manager's decisions within a mandate. A valid benchmark is specified in advance, appropriate, measurable, unambiguous, reflective of current investment opinions, accountable and investable. Know the Sortino ratio, appraisal ratio, upside and downside capture, maximum drawdown and drawdown duration, and their limitations.
- How it shows up
- An attribution table or set of return statistics is given and you identify the source of excess return, or compute the measure that matches the investor's concern.
- Common confusion
- Crediting the manager for an allocation effect that came from a policy the asset owner set.
3.2 Investment Manager SelectionDue diligence, Type I and II errors, style analysis, vehicles, contracts and fees.
- What it means
- This module covers the manager selection process (universe, quantitative and qualitative screening, due diligence on investment and operations), errors in hiring and firing decisions, returns-based and holdings-based style analysis, capture and drawdown measures, behavioral factors in team decisions, pooled versus separate accounts, contract types and performance-based fees.
- What to understand
- With the null hypothesis that a manager has no skill, a Type I error is hiring or keeping a manager without skill and a Type II error is not hiring or firing a manager who has skill. Returns-based style analysis is cheap and quick but can mislead when holdings change; holdings-based analysis is more precise but depends on snapshot data. Performance-based fees can align interests but add complexity; hurdles, high-water marks, caps and clawbacks shape the incentives.
- How it shows up
- A consultant's decision is described and you name the error risked, or a fee schedule is given and you compute the fee for a stated return.
- Common confusion
- Reversing Type I and Type II errors; anchor on the null hypothesis that the manager has no skill.
3.3 Overview of the Global Investment Performance StandardsGIPS objectives, compliance, return calculation, composites, presentation and verification.
- What it means
- This module covers the purpose and scope of GIPS, how a firm is defined and claims compliance, return calculation requirements for external cash flows, cash, fees and expenses, the recommended valuation hierarchy, composite construction and maintenance, presentation and reporting, portability of past performance and verification.
- What to understand
- Compliance is claimed by the firm as a whole and only when all applicable requirements are met; there is no partial compliance. Every actual, fee-paying, discretionary portfolio must be in at least one composite defined by mandate, objective or strategy, and portfolios may be moved between composites only when the client's mandate changes, with history staying in the original composite. Verification is performed firm-wide by an independent third party and is recommended rather than required.
- How it shows up
- A firm's composite, discretion or performance-linking practice is described and you decide whether it meets GIPS.
- Common confusion
- Believing verification certifies the accuracy of a specific composite's returns; it tests firm-wide composite construction and policies.
High-yield concepts
The terms, rules and formulas this domain turns into questions most often.
-
Measurement, attribution and appraisal
Measurement computes the return, attribution explains the difference from the benchmark, and appraisal judges whether that difference reflects skill.
- Remember
- Each answers a different question, and appraisal needs enough history and a valid benchmark to mean anything.
- Tested as
- Classifying what a report or analysis actually does.
- Trap
- Treating a large positive attribution effect as proof of skill.
-
Macro vs. micro attribution
Macro attribution measures the asset owner's decisions, such as strategic allocation and manager structure; micro attribution measures a manager's decisions within one mandate.
- Remember
- Assign each decision to whoever made it before crediting skill.
- Tested as
- Owner-versus-manager questions.
- Trap
- Crediting a manager with a policy allocation the owner set.
-
Brinson-style equity attribution
Splits active return into allocation (over- or underweighting segments), selection (picking securities within segments) and sometimes interaction.
- Remember
- Allocation uses the weight difference times benchmark segment performance; selection uses benchmark weights times the return difference within the segment.
- Tested as
- Interpreting or computing an effect from a table.
- Trap
- Swapping portfolio and benchmark weights in the selection effect.
-
Fixed-income attribution
Breaks bond returns into effects such as yield-curve level, slope and shape positioning, sector and credit allocation, and security selection.
- Remember
- Curve and duration effects reflect rate decisions; spread effects reflect credit decisions.
- Tested as
- Reading fixed-income attribution output.
- Trap
- Attributing a duration-driven gain to security selection.
-
Benchmark quality criteria
A valid benchmark is specified in advance, appropriate, measurable, unambiguous, reflective of current investment opinions, accountable and investable.
- Remember
- Misspecification distorts both attribution and appraisal; style drift against a poor benchmark can look like skill.
- Tested as
- Evaluating a proposed benchmark.
- Trap
- Accepting a peer-group median, which is not specified in advance or investable.
-
Sortino ratio
Portfolio return minus a minimum acceptable return, divided by downside deviation below that return.
- Remember
- It penalizes only harmful volatility, so it suits investors focused on shortfall or with asymmetric return profiles.
- Tested as
- Direct calculation and choice of measure.
- Trap
- Dividing by total standard deviation, which turns it into a Sharpe-style ratio.
-
Appraisal ratio
Alpha divided by the standard deviation of residual (nonsystematic) returns.
- Remember
- It measures alpha per unit of active, unsystematic risk and is useful for comparing active managers.
- Tested as
- Ranking managers on skill per unit of residual risk.
- Trap
- Using total risk in the denominator.
-
Capture ratios and drawdown
Upside capture is the share of benchmark gains captured in up markets; downside capture is the share of benchmark losses suffered; the capture ratio divides upside by downside.
- Remember
- Maximum drawdown is the largest peak-to-trough loss, and drawdown duration measures how long recovery takes.
- Tested as
- Evaluating managers for loss-averse clients.
- Trap
- Reading a downside capture above 100% as protective.
-
Type I vs. Type II error
With the null hypothesis that a manager has no skill, Type I is hiring or keeping an unskilled manager, and Type II is passing on or firing a skilled one.
- Remember
- Short track records and noisy returns make both errors likely; reducing one tends to raise the other.
- Tested as
- Naming the error in a hiring or firing decision.
- Trap
- Reversing the labels by forgetting the null hypothesis.
-
Style analysis
Returns-based analysis infers style by regressing returns on style indexes; holdings-based analysis classifies actual holdings.
- Remember
- Returns-based is cheap and consistent but backward-looking; holdings-based is more accurate at a point in time but can miss intra-period trading.
- Tested as
- Detecting style drift during due diligence.
- Trap
- Concluding that a manager changed style from returns-based output alone without checking holdings.
-
GIPS firm-wide compliance
Compliance is claimed by a firm, defined as an investment firm, subsidiary or division held out to clients as a distinct business entity, and only when all applicable requirements are met.
- Remember
- No partial compliance claims, and all actual fee-paying discretionary portfolios must be in at least one composite.
- Tested as
- Compliance judgment questions.
- Trap
- Claiming compliance for one strategy or composite.
-
Composite maintenance rules
Composites are defined by mandate, objective or strategy and must include all qualifying portfolios.
- Remember
- Move a portfolio only on a documented change in client mandate, keep its history in the original composite, include new portfolios on a consistent policy, and keep terminated portfolios' history through their last full period.
- Tested as
- Composite construction scenarios.
- Trap
- Removing a terminated portfolio's past returns because it underperformed.
-
GIPS verification
An independent third-party review of a firm's composite and pooled-fund policies and procedures on a firm-wide basis.
- Remember
- Verification is recommended, not required, and does not certify a specific composite's performance.
- Tested as
- Interpreting what a verification claim means.
- Trap
- Thinking verification can be done for one composite only.
Side by side
Concepts candidates mix up, compared on the point that decides the answer.
Sharpe vs. Sortino vs. appraisal ratio
| Concept | Meaning | Key difference | Common trap |
|---|---|---|---|
| Sharpe ratio | Excess return over the risk-free rate per unit of total volatility | Penalizes upside and downside volatility equally | Using it for strategies with skewed or option-like returns |
| Sortino ratio | Return above a minimum acceptable return per unit of downside deviation | Penalizes only volatility below the threshold | Using total standard deviation in the denominator |
| Appraisal ratio | Alpha per unit of residual (nonsystematic) risk | Focuses on active skill relative to the risk taken to earn it | Comparing it with ratios built on total risk |
Type I vs. Type II error
| Concept | Meaning | Key difference | Common trap |
|---|---|---|---|
| Type I error | Hiring or retaining a manager who has no skill | Rejects a true null hypothesis of no skill | Labeling the firing of a skilled manager as Type I |
| Type II error | Not hiring or firing a manager who does have skill | Fails to reject a false null hypothesis of no skill | Thinking a short bad run proves no skill |
Returns-based vs. holdings-based style analysis
| Concept | Meaning | Key difference | Common trap |
|---|---|---|---|
| Returns-based | Regresses manager returns on style index returns | Needs only return data; cheap and consistent over time | Missing style changes hidden by averaging |
| Holdings-based | Classifies the actual securities held | More precise at a point in time but needs holdings data | Missing window dressing or trading between snapshots |
Exam traps
Each trap is an answer that looks right. Learn the clue that should change your mind.
-
Positive excess return proves skill
- What candidates pick
- Conclude the manager is skilled because the portfolio beat the benchmark.
- Why it looks right
- Outperformance is the outcome everyone wants.
- Why it's wrong
- Appraisal requires risk adjustment, a valid benchmark and enough history; excess return from style drift or leverage against a poor benchmark is not skill.
Clue to watch for The benchmark does not match the manager's style or the period is short.
-
Wrong appraisal measure
- What candidates pick
- The Sharpe ratio for a client whose concern is losses below a required return.
- Why it looks right
- It is the most familiar risk-adjusted measure.
- Why it's wrong
- The client's concern is downside shortfall, so the Sortino ratio, downside capture or maximum drawdown answers the question directly.
Clue to watch for The vignette states a minimum acceptable return or emphasizes losses.
-
Firing after a short slump
- What candidates pick
- Replace a manager after one year of underperformance.
- Why it looks right
- It looks like decisive governance.
- Why it's wrong
- Short periods of noisy returns cannot separate skill from luck, so firing risks a Type II error and adds transition costs; the process review should drive the decision.
Clue to watch for The manager's process and team are unchanged and the period is brief.
-
Peer median as benchmark
- What candidates pick
- A peer-group median as the manager's benchmark.
- Why it looks right
- It shows relative standing against competitors.
- Why it's wrong
- A peer median is not specified in advance, not investable and suffers survivorship bias, so it fails benchmark quality tests.
Clue to watch for The proposed benchmark is a universe ranking rather than an index or custom portfolio.
-
Composite cherry-picking
- What candidates pick
- Accept a firm moving a strong portfolio into a new composite to improve its record.
- Why it looks right
- The portfolio does fit the new strategy description.
- Why it's wrong
- GIPS allows a move only on a documented change in the client's mandate, and the portfolio's history stays in the original composite.
Clue to watch for The move coincides with marketing a new composite, with no client instruction.
-
Verification as a performance guarantee
- What candidates pick
- Treat a verified firm's composite returns as certified accurate.
- Why it looks right
- Verification sounds like an audit of results.
- Why it's wrong
- Verification tests firm-wide policies and procedures for composite construction and calculation; it does not certify any specific composite's returns.
Clue to watch for The marketing says returns are verified for one composite.
Answer framework
For a performance vignette, decide who made each decision before judging whether it added value.
- Confirm the benchmark is valid: specified in advance, appropriate to the style, measurable, unambiguous, investable and accepted by the manager.
- Split total excess return into owner decisions (macro attribution) and manager decisions (micro attribution) before crediting anyone.
- Read the attribution output to find whether allocation, selection, duration, curve or credit choices drove the result.
- Choose the appraisal measure that fits the stated concern: Sortino or drawdown for downside, appraisal ratio for active skill, capture ratios for asymmetry.
- For selection or GIPS questions, test the decision against the specific rule or error type, naming the null hypothesis or the composite requirement involved.
Practice question
Want more like this? The CFA Level 3 practice test has questions for every domain.
Checklist
0 of 13
Tick each one you can do without notes. Your ticks stay in this browser.
Study strategy
Start with Portfolio Performance Evaluation, because its vocabulary (measurement, attribution, appraisal, macro and micro attribution) underpins the other two modules. Practice reading attribution tables until you can name the driving decision in seconds, then drill the appraisal measures with the same data set so you see how each ratio changes the ranking of managers.
Next, cover Investment Manager Selection. The high-priority items are Type I and Type II errors anchored on the no-skill null hypothesis, style analysis for detecting drift, and the incentive effects of performance-based fee features such as hurdles, high-water marks and clawbacks. Candidates commonly underestimate the qualitative due diligence content, including operational risk and behavioral factors in investment committees.
Finish with GIPS. Learn the rules as a short list of must-haves (firm-wide compliance, composite inclusion, mandate-based switching, portability conditions, verification scope) and practice yes-or-no compliance judgments. Afterwards, attempt mixed item sets that combine attribution and appraisal, and write two-sentence evaluations of benchmarks, then return to the pathway topic to see these tools applied to specialist mandates.
FAQ
How much of CFA Level 3 is Performance Measurement?
CFA Institute lists Performance Measurement at 5-10% of the Level 3 exam, the smallest core topic. It has three learning modules: Portfolio Performance Evaluation, Investment Manager Selection and an overview of GIPS.
Which ratios should I be able to calculate?
The Sortino ratio, the appraisal ratio, upside and downside capture and the capture ratio, maximum drawdown and drawdown duration, alongside the Sharpe and information ratios you already know. Know which investor concern each one addresses.
How do I avoid mixing up Type I and Type II errors?
Start from the null hypothesis that the manager has no skill. Hiring or keeping an unskilled manager rejects a true null (Type I); passing on or firing a skilled manager fails to reject a false null (Type II).
Is GIPS tested in the same depth as at earlier levels?
Level 3 goes further, covering return calculation rules, composite construction and maintenance, presentation and reporting, portability of past performance and verification. Expect compliance judgments on specific firm practices.
Why does benchmark quality matter so much?
Attribution and appraisal both compare results with the benchmark, so a benchmark that does not match the manager's style can make style drift look like skill or hide genuine skill. The quality criteria give you a checklist for spotting this.
What is the difference between macro and micro attribution?
Macro attribution covers the asset owner's decisions, such as the strategic allocation and choice of managers, while micro attribution covers a single manager's decisions within a mandate. Value should be credited to whoever made the decision.
How does this topic connect to the pathways?
Each pathway's specialist strategies are ultimately judged with these tools, from active equity and credit managers in Portfolio Management to private fund performance metrics in Private Markets and investment program evaluation in Private Wealth.
Other Finance certifications
Where the same subject shows up in other exams. Cards marked "Same subject" open the matching domain.
Test your Domain 3 knowledge
You have covered attribution, benchmark quality, appraisal measures, manager selection and GIPS. Test whether you can judge manager skill and compliance accurately under exam timing.
Keep going
- Training Training Built on your Weak areas
- Practice test Questions by domain, with explanations
- Mock exam Full length, timed like the real exam
- Readiness test Find your weakest domain in minutes
- Cheat sheet Key terms and formulas on one page
- Ask the CFA Level 3 AI tutor Get any concept explained your way