FRM Exam Part II · Portfolio Performance Evaluation

Benchmarking and Style Analysis for FRM Part II

Updated 11 October 2026 · Fact-checked

A benchmark is the yardstick you judge a manager against. A good one is unambiguous, investable, measurable, appropriate, specified in advance and owned by the manager. Returns-based style analysis regresses fund returns on style index returns, with non-negative weights summing to 100%, to find the manager's effective style mix.

Understand Benchmarking and Style Analysis

A benchmark is a reference portfolio used to judge a manager's performance. Without one, a 9% return tells you nothing. If the market returned 14%, the manager lost value. If it returned 3%, the manager added value. The benchmark turns a raw return into an active return.

A valid benchmark has recognised properties. It is unambiguous: the holdings and weights are clearly named. It is investable: you could hold it as a passive alternative. It is measurable: its return can be calculated often and accurately. It is appropriate: it matches the manager's investment style and mandate. It is specified in advance: set before the evaluation period starts. It reflects current investment opinions: the manager knows the securities in it. And the manager accepts accountability for it.

Returns-based style analysis (Sharpe) finds a manager's style without looking at holdings. You regress the fund's returns on the returns of a set of style indices, such as large-cap value, large-cap growth, small-cap and bonds. The weights are constrained to be non-negative and to sum to 100%. The fitted mix is the effective style mix. It is a custom benchmark built from the fund's own behaviour.

The fit is measured by R². The part of fund variance explained by style is the style component. The rest is the selection component, which reflects security selection and timing. The difference between fund return and the style-mix return is the manager's selection return.

A custom benchmark from style analysis is only as good as the data. It needs enough observations, sensible indices that are not highly correlated, and checks that the style stays stable over time. It also does not replace a holdings-based review.

Key formulas to remember

Returns-based style regression: R_fund,t = b1·F1,t + b2·F2,t + … + bn·Fn,t + e_t; F are style index returns. Constraints: every b ≥ 0 and Σb = 1 (100%). e_t is the selection residual.
Style R²: R² = 1 − Var(e) ÷ Var(R_fund); Share of fund return variance explained by the style mix. The remainder is selection.
Selection return: Selection return = R_fund − Σ(b_i × F_i); Return not explained by style. It is the active return versus the style benchmark.
Active return: Active return = R_portfolio − R_benchmark; Positive means the manager beat the benchmark in that period.
Properties of a valid benchmark: Unambiguous, investable, measurable, appropriate, specified in advance, reflects current investment opinions, accountable; Know each property and what breaks it.

How to solve Benchmarking and Style Analysis questions

Use this method for any benchmark or style analysis question.

1Identify what is asked: a benchmark property, a style analysis interpretation, or a calculation of active or selection return.
2Read the mandate. Note the asset class, region, size, style and constraints, because 'appropriate' depends on them.
3For benchmark validity, test the candidate against each property and find the one that fails.
4For style analysis, check the constraints: weights non-negative and summing to 100%. Read the weights as the effective style mix.
5Interpret R². High R² means style explains most of the return. Low R² means selection, timing or a missing style index.
6For calculations, compute the style-mix return as Σ weight × index return, then subtract it from the fund return.
7Compare the style mix to the stated mandate. A mismatch signals style drift or a poor benchmark.
8State the conclusion in one sentence and pick the option that matches it.

Quickest way: Property check and weighted-sum shortcut

When to use it: Use when an MCQ gives a benchmark description or a set of style weights and returns.

For a benchmark description, scan for the missing property: unclear holdings (ambiguous), cannot be traded (not investable), set after the fact (not in advance), wrong style (not appropriate).
For style numbers, confirm weights sum to 100% before anything else.
Multiply each weight by its index return and add. That is the benchmark return.
Subtract from the fund return to get the selection return.
Eliminate options with wrong sign or with arithmetic on unweighted returns.

Common mistakes in Benchmarking and Style Analysis

Treating any popular market index as an appropriate benchmark.
Popularity feels like validity.
Fix: Check that the index matches the manager's style, size and region. A large-cap index is wrong for a small-cap value fund.
Saying a benchmark can be chosen after seeing results.
Students confuse reviewing a benchmark with picking one retrospectively.
Fix: A valid benchmark is specified in advance. Choosing it afterwards invites cherry-picking.
Allowing negative weights in returns-based style analysis.
Ordinary regression allows them.
Fix: Sharpe's style analysis constrains weights to be non-negative and sum to 100%, so they read as a long-only mix.
Reading a low R² as proof of skill.
Low R² means unexplained return, which sounds like alpha.
Fix: Low R² only means style indices do not explain returns. It may be selection, timing, leverage, derivatives or missing indices.
Confusing investable with measurable.
Both sound like practical properties.
Fix: Investable means you could hold it passively. Measurable means its return can be calculated reliably and often.
Adding index returns without weighting them.
Rushing under time pressure.
Fix: Always multiply each index return by its style weight before summing.

Worked examples

Example 1

A fund's style analysis gives weights of 60% large-cap value, 30% small-cap and 10% bonds. Index returns for the year were 12%, 18% and 4% respectively. The fund returned 15.5%. What is the selection return?

Show the solution

Check the weights: 60% + 30% + 10% = 100%.
Style-mix return = 0.60 × 12% + 0.30 × 18% + 0.10 × 4%.
= 7.2% + 5.4% + 0.4% = 13.0%.
Selection return = 15.5% − 13.0% = 2.5%.

Answer: The selection return is 2.5%. The fund beat its style-mix benchmark by that amount.

Example 2

A manager of a global equity fund is judged against a benchmark chosen by the committee after the year ended, because it was the index with the lowest return. The benchmark's constituents are published only quarterly. Which properties of a valid benchmark are violated?

Show the solution

The benchmark was chosen after the period, so it was not specified in advance.
Choosing the lowest-return index also fails appropriateness, since selection was based on outcome, not mandate.
Constituents published only quarterly weaken the measurable and unambiguous properties, because returns cannot be calculated reliably between publication dates.
The manager had no say in it beforehand, so accountability is also undermined.

Answer: The benchmark was not specified in advance and is not appropriate. Its infrequent disclosure also weakens measurability. It is not a valid benchmark.

Exam tips

Questions often describe a flawed benchmark in a scenario. Name the single property that fails.
Remember the two style analysis constraints: non-negative weights and weights summing to 100%.
Interpret R² as the share of variance explained by style, with the rest attributed to selection.
Watch for style drift. A style mix that differs from the stated mandate is a red flag.
Do the weighted-sum arithmetic carefully. Distractors usually come from unweighted or mis-weighted sums.

Practice questions from Portfolio Performance Evaluation

Benchmarking and Style Analysis in other exams

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

Benchmarking and Style Analysis: frequently asked questions

What are the characteristics of a good benchmark?

It is unambiguous, investable, measurable, appropriate to the mandate, specified in advance, reflects current investment opinions and is accepted by the manager. A benchmark that fails any of these gives a weaker judgement of performance.

What is returns-based style analysis?

It regresses a fund's returns on the returns of style indices, with non-negative weights summing to 100%. The weights show the effective style mix. It needs only return data, not holdings.

How do I choose a benchmark for portfolio evaluation?

Start from the mandate: asset class, region, size and style. Pick an index or custom blend that matches it and that the manager could actually hold. Fix it before the evaluation period begins.

What does a low R² in style analysis mean?

It means the chosen style indices explain little of the fund's return variation. The cause could be active selection, timing, leverage or a missing index. It does not by itself prove skill.