Browse Certification Practice Tests by Exam Family

CISI IM: Data Analysis

Try 10 focused CISI IM questions on Data Analysis, with answers and explanations, then continue with Securities Prep.

On this page

Open the matching Securities Prep practice page for timed mocks, topic drills, progress tracking, explanations, and full practice.

Topic snapshot

FieldDetail
Exam routeCISI IM
IssuerCISI
Topic areaData Analysis
Blueprint weight13%
Page purposeFocused sample questions before returning to mixed practice

How to use this topic drill

Use this page to isolate Data Analysis for CISI IM. Work through the 10 questions first, then review the explanations and return to mixed practice in Securities Prep.

PassWhat to doWhat to record
First attemptAnswer without checking the explanation first.The fact, rule, calculation, or judgment point that controlled your answer.
ReviewRead the explanation even when you were correct.Why the best answer is stronger than the closest distractor.
RepairRepeat only missed or uncertain items after a short break.The pattern behind misses, not the answer letter.
TransferReturn to mixed practice once the topic feels stable.Whether the same skill holds up when the topic is no longer obvious.

Blueprint context: 13% of the practice outline. A focused topic score can overstate readiness if you recognize the pattern too quickly, so use it as repair work before timed mixed sets.

Sample questions

These questions are original Securities Prep practice items aligned to this topic area. They are designed for self-assessment and are not official exam questions.

Question 1

Topic: Data Analysis

An investment analyst is building a portfolio dashboard and must classify each field correctly before choosing summary statistics and charts. Which classification best fits daily portfolio return, number of separate securities held, and investment style (value, growth, or core)?

  • A. Return continuous; holdings categorical; style discrete
  • B. Return continuous; holdings discrete; style categorical
  • C. Return discrete; holdings continuous; style categorical
  • D. Return categorical; holdings discrete; style continuous

Best answer: B

What this tests: Data Analysis

Explanation: The correct classification depends on how each variable is measured. Daily portfolio return can take fractional values, so it is continuous; number of securities held is a whole-number count, so it is discrete; and investment style is a named grouping, so it is categorical.

The core concept is to match each data item to its measurement type before selecting charts, summaries, or statistical methods. Daily portfolio return is continuous because it can take any value within a range, including decimals such as 0.18% or -0.73%. Number of separate securities held is discrete because it is a count and is normally expressed in whole numbers. Investment style is categorical because terms such as value, growth, and core are labels, not measured quantities.

  • Continuous data: measured on a scale
  • Discrete data: counted items
  • Categorical data: class labels

A common mistake is to treat any numeric-looking field as continuous, but counts and labels need different handling.

  • Treating return as discrete and holdings as continuous reverses a measured value and a count.
  • Treating return as categorical mistakes a numeric percentage for a label.
  • Treating style as continuous or discrete confuses named groups with quantities.
  • Treating holdings as categorical ignores that it is still a numeric count.

Portfolio return is measured on a continuum, holdings are counted in whole numbers, and style is a label.


Question 2

Topic: Data Analysis

A data set of four returns has a mean of 5%, and the sum of squared deviations from the mean is 24. Which pair of variance values is correct?

  • A. Population variance = 6; sample variance = 8
  • B. Population variance = 8; sample variance = 8
  • C. Population variance = 8; sample variance = 6
  • D. Population variance = 6; sample variance = 6

Best answer: A

What this tests: Data Analysis

Explanation: Population variance divides the sum of squared deviations by the number of observations, while sample variance divides by one less. With four observations and a squared-deviation total of 24, the values are 6 and 8 respectively.

Variance measures dispersion around the mean. If the data represent the whole population, divide the sum of squared deviations by n. If the data are treated as a sample, divide by n - 1 because one degree of freedom is used in estimating the mean. Here, n = 4 and the sum of squared deviations is 24.

  • Population variance: 24/4 = 6
  • Sample variance: 24/3 = 8

So the correct pair is 6 for the population and 8 for the sample. The main trap is reversing the two denominators.

  • Reversing 8 and 6 swaps the population denominator with the sample denominator.
  • Giving 6 for both ignores the n - 1 adjustment required for sample variance.
  • Giving 8 for both wrongly applies the sample denominator to the whole population.

Population variance uses n = 4, while sample variance uses n - 1 = 3, so 24/4 = 6 and 24/3 = 8.


Question 3

Topic: Data Analysis

To assess the relationship between a fund’s monthly returns and its benchmark’s monthly returns over the last three years, which presentation method is most appropriate?

  • A. Table of monthly returns
  • B. Scatter plot
  • C. Line chart
  • D. Histogram

Best answer: B

What this tests: Data Analysis

Explanation: A scatter plot is best when the aim is to examine the relationship between two variables using paired observations. Here, each point can represent one month’s fund return against the benchmark return, making correlation, clustering, and outliers visible.

The core concept is matching the chart type to the data question being asked. When you want to see how two return series relate to each other, a scatter plot is most suitable because it plots paired observations on two axes. That makes it easier to judge whether fund returns tend to move in line with benchmark returns, and whether the relationship appears strong, weak, or unstable.

A line chart is better for showing how returns change through time. A histogram is better for showing the distribution or frequency of returns. A table gives exact figures, but it is less effective for visualising the relationship between two variables.

The key distinction is relationship between two series, not trend over time or frequency of outcomes.

  • Line chart: useful for time-series trends, but less suitable for judging the relationship between two variables from paired observations.
  • Histogram: shows the distribution of one variable, such as the frequency of monthly returns, not how two series move relative to each other.
  • Table: provides precise data values, but patterns of association are harder to identify visually than on a scatter plot.

A scatter plot shows paired return observations and makes the strength and direction of the relationship easiest to see.


Question 4

Topic: Data Analysis

An analyst applies a three-factor model to a portfolio’s quarterly excess return:

\[ R_p - R_f = \alpha + \beta_M F_M + \beta_S F_S + \beta_V F_V \]

Exhibit:

  • \(\beta_M = 1.1\)
  • \(\beta_S = -0.4\)
  • \(\beta_V = 0.6\)
  • Quarter factor returns: \(F_M = 4.0\%\), \(F_S = -1.5\%\), \(F_V = 2.0\%\)
  • Actual portfolio excess return: \(5.9\%\)

Which interpretation is most accurate?

  • A. Alpha is about -0.3% under this model; realised return lagged factor-implied return.
  • B. Alpha is zero because multi-factor models fully explain excess returns.
  • C. Alpha is about -1.5% under this model; the size effect is counted twice.
  • D. Alpha is about +0.3% under this model; realised return beat factor-implied return.

Best answer: A

What this tests: Data Analysis

Explanation: A multi-factor model decomposes excess return into factor-driven return plus alpha. Here the factor contribution is \(1.1\times4.0\% + (-0.4)\times(-1.5\%) + 0.6\times2.0\% = 6.2\%\), so the portfolio’s actual excess return of 5.9% implies an alpha of -0.3%.

The core use of a multi-factor model is to separate return explained by systematic factor exposures from any residual return, or alpha. In this case, the market factor contributes 4.4%, the size factor contributes 0.6% because a negative exposure multiplied by a negative factor return is positive, and the value factor contributes 1.2%. That gives a total factor-implied excess return of 6.2%.

  • Factor-implied return = 4.4% + 0.6% + 1.2% = 6.2%
  • Alpha = actual excess return - factor-implied return
  • Alpha = 5.9% - 6.2% = -0.3%

So the portfolio underperformed what its factor exposures would have predicted. The key limitation is that this conclusion is model-dependent: a different factor set or estimation period could produce a different alpha.

  • Sign error: a positive 0.3% alpha comes from reversing the subtraction; alpha is actual minus factor-implied return.
  • Sequence error: the negative size beta and negative size factor return create a positive contribution, not a double deduction.
  • Overclaim: multi-factor models do not automatically explain all excess return; unexplained residuals remain and depend on model specification.

The factor-implied excess return is 6.2%, so alpha is \(5.9\% - 6.2\% = -0.3\%\).


Question 5

Topic: Data Analysis

An investment analyst drafts the following quarterly report plan.

Report itemPlanned display
Monthly benchmark returns, last 36 monthsHistogram
Distribution of bond trade sizes, 300 deals this quarterLine chart
25 shares: dividend yield and price-to-bookScatter plot
Portfolio weights at 31 MarchLine chart

Which planned display should be kept unchanged?

  • A. The histogram for 36 months of monthly benchmark returns
  • B. The line chart for portfolio weights at 31 March
  • C. The line chart for the distribution of 300 bond trade sizes
  • D. The scatter plot for 25 shares’ dividend yield and price-to-book

Best answer: D

What this tests: Data Analysis

Explanation: A scatter plot is used to show the relationship between two numerical variables across a set of observations. Dividend yield and price-to-book for 25 shares fit that purpose, while the other items are a time series, a frequency distribution, and a point-in-time snapshot.

The key skill is matching the display to the structure of the data. A scatter plot is the best choice when the objective is to see whether two continuous variables move together across observations, such as dividend yield and price-to-book across 25 shares. Monthly benchmark returns over 36 months are time-series data, so a line chart would normally show trend more clearly than a histogram. A distribution of trade sizes is better shown with a histogram because the aim is to show frequency by size range. Portfolio weights at a single date are point-in-time values, so a table is usually clearer when exact percentages matter.

The correct choice is the only pairing that already matches the data type and analytical purpose.

  • Time series confusion: Monthly benchmark returns are ordered through time, so a line chart is normally better than a histogram for showing trend.
  • Distribution confusion: Trade sizes across 300 deals are frequency data, so a histogram is more suitable than a line chart.
  • Snapshot confusion: Portfolio weights at one date are not a trend series; a table is usually clearer for exact point-in-time allocations.

A scatter plot is appropriate for examining the relationship between two quantitative variables across multiple securities.


Question 6

Topic: Data Analysis

An investment committee is reviewing whether its manager benchmark matches the mandate.

Exhibit:

  • Mandate: balanced GBP portfolio
  • Strategic weights: UK equities 40%, overseas equities 20%, GBP investment-grade bonds 35%, cash 5%
  • Overseas assets are assessed on a GBP-hedged basis
  • Current benchmark in the performance report: FTSE All-Share

Which benchmark approach is most suitable for evaluating the manager against the mandate?

  • A. Use a peer-group median return to reflect manager flexibility.
  • B. Use a 40/20/35/5 composite of matching equity, bond and cash indices.
  • C. Use one global equity index to reflect the overseas holdings.
  • D. Keep FTSE All-Share because equities are the largest asset class.

Best answer: B

What this tests: Data Analysis

Explanation: A composite benchmark is more suitable when a portfolio mandate spans several asset classes or regions. Here, a single UK equity index ignores overseas equities, bonds, cash and the GBP-hedged measurement basis, so a weighted multi-index benchmark best reflects the manager’s remit.

A composite or synthetic benchmark combines relevant market indices in preset weights so that performance is judged against the actual mandate rather than one market segment. In the exhibit, the portfolio is not a pure UK equity mandate: it includes UK equities, overseas equities, GBP investment-grade bonds and cash, and the overseas sleeve is assessed on a GBP-hedged basis. FTSE All-Share represents only UK equities, so it cannot fairly support manager evaluation or attribution for the whole portfolio. The most appropriate benchmark is therefore a weighted blend of suitable equity, bond and cash indices aligned to the 40/20/35/5 strategic allocation, with the overseas component measured consistently with the hedged mandate. A single global equity index is closer than FTSE All-Share, but it still ignores the fixed-income and cash exposures.

  • Largest-slice trap: UK equities are only 40% of the mandate, so one UK equity index does not represent the full portfolio.
  • Overseas-only trap: a global equity index captures foreign shares better, but still misses 35% bonds and 5% cash.
  • Peer-group trap: a peer median is a comparator set, not a mandate-specific benchmark for attribution and manager assessment.

A weighted composite benchmark reflects the mandate’s multi-asset structure, unlike a single UK equity index.


Question 7

Topic: Data Analysis

A discretionary manager is preparing a two-year portfolio review. The portfolio returned +40% in year 1 and -20% in year 2. Rounded to one decimal place, which response best gives the average annual return to report if the aim is to show the portfolio’s compounded growth rate over the two years?

  • A. 20.0%, because the loss should simply be netted against the gain
  • B. 12.0%, because the cumulative two-year gain is the annual average
  • C. 5.8%, using the geometric mean for compounded multi-period returns
  • D. 10.0%, using the arithmetic mean for the annualised client report

Best answer: C

What this tests: Data Analysis

Explanation: The correct measure for average annual compounded growth over multiple periods is the geometric mean, not the arithmetic mean. Here, the portfolio grows by a factor of 1.40 and then 0.80, giving an overall factor of 1.12, so the annualised compounded return is 5.8%.

When a client report needs the annual rate that links the starting value to the ending value over more than one period, the correct return measure is the geometric mean. It captures compounding, whereas the arithmetic mean is just the simple average of periodic returns.

$$\begin{aligned} \text{Geometric mean} &= \left[(1+0.40)(1-0.20)\right]^{1/2}-1 \ &= (1.12)^{1/2}-1 \ &\approx 0.0583 = 5.8% \end{aligned}$$

The arithmetic mean is \((40\%-20\%)/2 = 10.0\%\), but that overstates the annualised growth rate because the 20% loss occurs on a larger capital base after the 40% gain. The key takeaway is that volatility causes the geometric mean to be lower than the arithmetic mean unless returns are identical each period.

  • Arithmetic mean trap: 10.0% is the simple average of the two yearly returns, but it does not represent compounded growth over time.
  • Total return trap: 12.0% is the two-year cumulative gain from 100 to 112, not the annualised average return.
  • Netting trap: Treating +40% and -20% as a simple 20% net change ignores that percentage gains and losses apply to different capital amounts.

For multi-period compounded growth, the geometric mean is \(\sqrt{1.40 \times 0.80}-1=5.8\%\), and it is lower than the arithmetic mean because volatility reduces compound growth.


Question 8

Topic: Data Analysis

A selector reviews 10 UK smaller-companies funds for the quarter. Nine funds returned between 4.8% and 6.2%, but one fund returned 29.0% after a takeover bid. The peer-group mean return is 7.8% and the median return is 5.6%. A candidate manager returned 6.0%. Which interpretation is best?

  • A. The manager lagged a typical peer, because 6.0% is below the mean and all valid returns should drive the comparison equally.
  • B. The manager matched a typical peer, because mean and median are interchangeable when the peer group is small.
  • C. The manager beat a typical peer, because 6.0% exceeds the median and the mean is distorted upward.
  • D. No useful peer comparison is possible, because one extreme return makes central-tendency measures unreliable.

Best answer: C

What this tests: Data Analysis

Explanation: The peer-group distribution is positively skewed by one exceptional 29.0% return, so the mean overstates the return achieved by most funds. A 6.0% result is above the 5.6% median, so the manager has outperformed a typical peer even though it is below the 7.8% mean.

When a data set contains a genuine but extreme outlier, the most useful central-tendency measure depends on the question being asked. Here, the question is about the typical peer outcome. The mean includes every observation, but one takeover-driven return of 29.0% pulls it well above the tight cluster of the other funds. The median is less sensitive to that skew and therefore gives a better picture of what a typical fund delivered.

Since the candidate manager returned 6.0%, it is above the peer-group median of 5.6%. That supports the conclusion that the manager beat a typical peer for the quarter. The mean still describes the overall average, but it is not the best guide to the typical manager in this skewed distribution.

  • Mean trap: The mean is affected by the single 29.0% return, so using it alone overstates the typical peer outcome.
  • Interchangeability trap: Mean and median are not interchangeable when returns are skewed; the data shape matters.
  • Overreaction trap: One valid extreme return does not make comparison impossible; it simply makes the median more informative here.

The median better represents the typical peer here because a single 29.0% outlier has pulled the mean above the main cluster of returns.


Question 9

Topic: Data Analysis

A UK balanced fund manager is being reviewed against a benchmark. The pension scheme made a large additional contribution halfway through the quarter, just before a broad market fall. The trustees want a return measure that isolates the manager’s investment performance from the timing of this external cash flow. Which measure is most appropriate?

  • A. Money-weighted rate of return
  • B. Real return after inflation
  • C. Time-weighted rate of return
  • D. Simple holding period return

Best answer: C

What this tests: Data Analysis

Explanation: The key issue is manager appraisal when an external cash flow occurred at a potentially distortive time. Time-weighted return removes the impact of the client’s contribution timing by measuring and linking returns across sub-periods, so it is the best measure of the manager’s underlying performance.

When the objective is to assess manager skill, the preferred measure is the time-weighted rate of return. In this scenario, the large mid-quarter contribution was an external cash flow decided by the client, not by the manager. A time-weighted approach breaks the period around that cash flow and compounds the sub-period returns, so the result is not distorted by the size or timing of the contribution.

Money-weighted return, by contrast, gives weight to when cash enters or leaves the portfolio, so it reflects the investor’s experience rather than the manager’s pure investment performance. That makes it useful for investor-specific outcomes, but not the best tool for manager comparison against a benchmark. The decisive point is that external cash-flow timing should be neutralised here.

  • Money-weighted return: plausible because it includes all cash flows, but that is exactly why it is unsuitable for isolating manager skill.
  • Simple holding period return: too crude because it does not properly adjust for the large contribution made during the measurement period.
  • Real return after inflation: useful for purchasing-power analysis, but inflation is not the issue being tested in this manager-review context.

It links sub-period returns and neutralises the effect of external cash flows, so it is the standard measure for assessing manager skill.


Question 10

Topic: Data Analysis

A portfolio analyst is comparing two UK equity mandates over the same 12-month period.

Exhibit:

PortfolioTotal returnBeta
Portfolio A9%0.8
Portfolio B11%1.4

Risk-free rate for the period: 3%

Which interpretation is best supported by the exhibit?

  • A. Portfolio B ranks higher on Treynor because it has the higher beta.
  • B. Portfolio B delivered more excess return because its total return was higher.
  • C. Treynor cannot be used here because standard deviation is not shown.
  • D. Portfolio A delivered more excess return per unit of systematic risk.

Best answer: D

What this tests: Data Analysis

Explanation: The Treynor ratio is excess return divided by beta, so it measures return per unit of systematic risk. Portfolio A has a Treynor ratio of 7.5, while Portfolio B is about 5.7, so Portfolio A used market risk more efficiently over the period.

The core concept is that the Treynor ratio evaluates excess return relative to systematic risk, using beta rather than total volatility. Because both portfolios are measured over the same period and use the same risk-free rate, they can be compared directly.

\[ \begin{aligned} T_A &= \frac{9\%-3\%}{0.8} = 7.5 \\ T_B &= \frac{11\%-3\%}{1.4} \approx 5.7 \end{aligned} \]

Although Portfolio B achieved the higher raw return, it took more market risk to generate that return. Portfolio A therefore produced the better Treynor outcome. The key takeaway is that Treynor rewards higher excess return only when it is earned efficiently relative to beta.

  • Higher return trap: The 11% portfolio is not automatically better on Treynor, because the ratio adjusts excess return for beta.
  • Sharpe confusion: Standard deviation is needed for the Sharpe ratio, not for the Treynor ratio.
  • Beta misunderstanding: A higher beta means more systematic risk; it does not by itself improve risk-adjusted performance.
  • Scope: The exhibit supports a Treynor comparison only, not a conclusion about every possible risk measure.

Its excess return divided by beta is higher: \((9\%-3\%)/0.8 = 7.5\), versus \((11\%-3\%)/1.4 \approx 5.7\).

Continue with full practice

Use the CISI IM Practice Test page for the full Securities Prep route, mixed-topic practice, timed mock exams, and explanations.

Open the matching Securities Prep practice route for timed mocks, topic drills, progress tracking, explanations, and full practice.

Free review resource

Use the full Securities Prep practice page above for the latest review links and practice route.

Revised on Thursday, May 14, 2026