Expect this module to appear as short calculation-and-interpretation questions. A stem may give a return series and ask for the average, spread, shape, co-movement, downside risk, or risk relative to average return. If you choose a familiar formula before identifying that target, a correct calculation can still answer the wrong question.
These outcomes form a chain. A return series gives you a typical result and its dispersion. Its shape explains why mean and median can separate. Paired series lead to covariance and correlation. A target or a mean changes the spread question to semi-deviation or coefficient of variation.
Read the image as a progression from observations to questions about returns. Begin with what a typical period looks like, then check how widely results vary and whether the distribution is asymmetric. When returns arrive in paired series, ask how they move together. A target return or average changes the focus again, because downside risk and risk per unit of average return answer different decisions.
Loading diagram…
Treating the arithmetic mean as the only meaningful average feels natural because it is the default average. It fails when the question concerns compounded wealth or strongly asymmetric returns. The clue is that the geometric mean or median may represent the question better. The calculator display below works through the four quarterly investor returns as decimals: first add the observations, then divide by four. The result is the typical one-quarter arithmetic return, not an annualized figure.
| Keys | Display |
|---|---|
| 2ND CLR TVM | 0.00 |
| -0.10 + 0.02 + 0.08 + 0.14 | 0.14 |
| ÷ 4 | 0.0350 |
For the numerical check, the observations are investor holding-period returns for four quarters, expressed as decimal returns per quarter. Treat them as a sample and do not annualize them unless the stem asks.
Measures of central tendency, location and dispersion - $\bar R=\sum R_t/n$ for a typical period; geometric mean compounds each $1+R_t$, while sample variance divides squared deviations by $n-1$ and population variance by $N$. The exam may ask for a typical or compounded return, middle or most frequent value, or spread. Match the operation to that wording. The trap is using $n$ for every variance or reporting the arithmetic mean for compounded wealth.
Covariance and correlation of returns - Sample covariance averages paired cross-products of deviations using $n-1$, and correlation divides it by both standard deviations, giving a unit-free value from $-1$ to $+1$. The exam may give paired same-period returns and ask about sign, strength, units, or comparison across assets. Read the requested scale carefully. The trap is treating covariance and correlation as interchangeable: covariance retains units, while correlation does not.
Skewness and kurtosis - Skewness uses cubed deviations to describe asymmetry; kurtosis uses fourth-power deviations to describe tail thickness. A long right tail generally pulls mean above median, while a long left tail generally puts mean below median. The exam may provide a histogram, tail description, or mean-median relation and ask for shape. The trap is reading positive skew as favorable performance; interpret tail direction, not the emotional meaning of the sign.
Semi-deviation and coefficient of variation - Semi-deviation squares only shortfalls below target $T$ (or the mean), averages them under the stated convention, then takes the square root; $CV=\operatorname{SD}(R)/\bar R$. The exam cues semi-deviation with a minimum acceptable return and CV with relative risk. The trap is substituting total standard deviation for downside dispersion, or overlooking that a negative mean can reverse CV's sign.
The first chart plots the four quarterly observations, so compare each point with the stated mean before deciding which returns fall below it. The following table makes that comparison explicit and preserves the decimal return scale. Read the two columns together: the return shows the observation, and the final column identifies its position relative to $0.0350$. This is the starting point for downside calculations.
The line plots each investor return in decimal form.
Loading diagram…
The plotted values make the two negative and two positive deviations visible, but the mean is the reference that determines which side each return occupies. The table records that classification explicitly. In a downside question, keep only observations below the stated target or mean before squaring shortfalls; in a question about total dispersion, retain observations on both sides. The same observations can therefore support different risk measures.
| Period | Return | Relative to mean |
|---|---|---|
| Q1 | -0.10 | Below $0.0350$ |
| Q2 | 0.02 | Below $0.0350$ |
| Q3 | 0.08 | Above $0.0350$ |
| Q4 | 0.14 | Above $0.0350$ |
Use this comparison when the stem asks which average fits the story. Read the measure column as the candidate operation and the decision cue as the stem's clue; the key is to choose before calculating. A typical-period question calls for a sum divided by observations, while unequal importance calls for stated weights. Compounded wealth requires multiplying growth factors. Sorting locates a median or mode, while tail-sensitive questions need a different measure.
| Measure | Operation | Decision cue |
|---|---|---|
| Arithmetic mean | Sum divided by observations | Typical one-period return |
| Weighted mean | Stated weights times returns | Unequal importance |
| Geometric mean | Compound each $1+R_t$ | Compounded wealth |
| Median | Middle sorted observation | Asymmetric returns |
| Mode | Most frequent observation | Repeated values |
These choices answer different descriptions of location, so do not let a familiar average dominate the wording. The calculator lines that follow retain the same four observations and display the arithmetic mean per quarter. Use that value as the reference point when interpreting the later comparison table; it is not a geometric return, a median, or a volatility estimate. Keeping those quantities distinct prevents a correct operation from answering the wrong question.
| Keys | Display |
|---|---|
| 2ND CLR TVM | 0.00 |
| -0.10 + 0.02 + 0.08 + 0.14 | 0.14 |
| ÷ 4 | 0.0350 |
The calculator result above is a worked arithmetic mean for the four quarterly returns. In an exam stem, first identify whether the requested result is a location measure, a spread measure, distribution shape, or co-movement. That classification determines which operation belongs next. Frequency and sample-versus-population wording also matter: preserve the stated period and denominator unless the question explicitly asks for a different basis.
| Question shape | What is given | What is asked and the trap |
|---|---|---|
| One-series calculation | Periodic returns, frequency, convention | Choose mean or spread. Mix frequency or denominator. |
| Shape interpretation | Tails, histogram, mean and median | Interpret skew or kurtosis. Confuse shape with performance. |
| Paired-return calculation | Paired returns, covariance, standard deviations | Compute co-movement. Treat covariance as unit-free. |
| Downside or relative risk | Target or mean, returns, standard deviation | Choose semi-deviation or coefficient of variation. Ignore target or mean sign. |
Start by identifying the requested object. Words such as typical one-period return, compounded wealth, middle, and most frequent point to the central-tendency choices. A return frequency in the stem stays with the answer unless annualization is requested. The sample or population label controls the variance and covariance denominator.
Tail language points to skewness or kurtosis. Two return columns observed in the same period point to covariance or correlation. A target or minimum acceptable return points to semi-deviation, while a mean and standard deviation with no target point to coefficient of variation. Treating correlation as interchangeable with covariance feels reasonable because both describe co-movement. The request for a unit-free comparison gives the difference away. Correlation is bounded between $-1$ and $+1$, while covariance retains units.
| Keys | Display |
|---|---|
| 2ND CLR TVM | 0.00 |
| -0.10 + 0.02 + 0.08 + 0.14 | 0.14 |
| ÷ 4 | 0.0350 |
After the exam-pattern table, use the short decision guide to translate the stem into a measure. “Typical” and “compounded” describe different return questions, and the period frequency stays attached to the answer unless annualization is requested. A sample label changes the variance or covariance denominator. These small wording cues often decide whether a calculation is relevant before any numbers are entered.
Practice
Using the quarterly investor returns -0.10, 0.02, 0.08, and 0.14, what is the arithmetic mean return per quarter?
Which statement best describes a positively skewed return distribution?
Which choice correctly evaluates correlation relative to covariance?
Which choice correctly distinguishes semi-deviation and coefficient of variation?
Use the requested object before touching the calculator.
Loading diagram…
Use the flow from left to right as a quick exam routine: identify the requested object, select the matching average or dispersion measure, then interpret shape or paired returns when those details appear. Downside analysis adds one extra check: establish whether the reference is a target or the mean, and inspect the mean's sign before using coefficient of variation. This order helps prevent a familiar formula from dictating the answer.
| Keys | Display |
|---|---|
| 2ND CLR TVM | 0.00 |
| -0.10 + 0.02 + 0.08 + 0.14 | 0.14 |
| ÷ 4 | 0.0350 |
The calculator display again confirms the arithmetic mean of $0.0350$ per quarter. The summary table below gathers the formulas and decision cues, so use it to distinguish location, full dispersion, co-movement, and downside risk. In particular, sample variance and covariance use $n-1$, whereas the population versions use $N$. Correlation rescales covariance, and the coefficient of variation scales standard deviation by the mean.
| Measure | Formula or decision | Decision cue |
|---|---|---|
| Arithmetic mean | $\bar R=\frac{\sum R_t}{n}$ | Sum periodic returns, divide observations |
| Weighted mean | Use stated weights for each $R_t$ | Unequal importance |
| Geometric mean | $\left[\prod(1+R_t)\right]^{1/n}-1$ | Compound before annualizing |
| Median or mode | Sort first, then middle or most frequent | Location without tail averaging |
| Range and mean absolute deviation | $\max(R_t)-\min(R_t)$, $\frac{\sum\lvert R_t-\bar R\rvert}{n}$ | Full spread versus absolute spread |
| Variance and standard deviation | Sample $s^2=\frac{\sum(R_t-\bar R)^2}{n-1}$, population $\sigma^2=\frac{\sum(R_t-\mu)^2}{N}$ | Square root variance for standard deviation |
| Skewness | Standardized third moment | Positive skew mean greater median, negative less |
| Kurtosis | Standardized fourth moment | Higher values signal fatter tails |
| Covariance | $\operatorname{Cov}(R,S)=\frac{\sum(R_t-\bar R)(S_t-\bar S)}{n-1}$ | Population uses $N$, units retained |
| Correlation | $\rho_{RS}=\frac{\operatorname{Cov}(R,S)}{\operatorname{SD}(R)\operatorname{SD}(S)}$ | Unit-free, between $-1$ and $+1$ |
| Semi-deviation | Below target or mean only, square and average | Downside spread, not both sides |
| Coefficient of variation | $CV=\frac{\operatorname{SD}(R)}{\bar R}$ | Check the mean sign |
Use the requested object before touching the calculator. Keep every return on the stated frequency, use $n-1$ only for a sample, and let the target decide which observations enter semi-deviation. A mean near zero or below zero can make coefficient of variation misleading. A periodic result is not annualized unless the question asks for it.
Signup to continue reading this lesson and unlimited access questions, flashcards, AI notes, and more
1. An investment pays USD 10,000 at each year-end for three years and USD 50,000 at the end of Year 4. At an annual discount rate of 8%, the calculated value of the investment today is closest to:
Answer: A. USD 62,522
USD 62,522 - Discount each payment from its own year-end: USD 10,000/1.08 + USD 10,000/1.08² + USD 10,000/1.08³ + USD 50,000/1.08⁴. - The sum is USD 62,522.46, so the value today is closest to USD 62,522. USD 58,803 - Discounting the entire USD 80,000 for four years gives USD 58,802.39, which is closest to USD 58,803. - That slip treats the first three USD 10,000 payments as though they arrive at the end of Year 4. USD 65,463 - Discounting the USD 50,000 payment for three years instead of four, while correctly discounting the other payments, gives USD 65,462.58. - The USD 50,000 arrives at the end of Year 4, so it requires four years of discounting.
2. For H0: μ ≤ 8% versus Ha: μ > 8%, a sample of 64 returns has mean 9.4% and standard deviation 4.0%. If the 5% critical t-value is 1.669, the analyst should:
Answer: A. Reject the null; the mean return exceeds 8%
Reject the null; the mean return exceeds 8% - The standard error is 4.0%/√64 = 0.5%, so the test statistic is (9.4% − 8.0%)/0.5% = 2.80. - Because 2.80 exceeds the upper-tail critical value of 1.669, reject H0 in favor of μ > 8%. Fail to reject the null; the mean return equals 8% - A test statistic of 2.80 exceeds 1.669, so failing to reject H0 is incorrect. - Even if the test had failed to reject H0, that would not establish that μ equals 8%. Reject the null; the mean return is below 8% - The sample mean of 9.4% is above 8%, giving a positive test statistic of 2.80. - The alternative is μ > 8%, so rejection supports a mean above 8%, not below it.
3. An analyst models continuously compounded returns as normally distributed, implying terminal asset prices are lognormally distributed. Which conclusion is most appropriate?
Answer: A. Zero probability of a negative terminal price
Zero probability of a negative terminal price - A terminal price is the positive initial price multiplied by the exponential of a continuously compounded return. - Because that exponential is strictly positive, a lognormally distributed terminal price cannot be negative. Equal probabilities of gains and losses - A gain occurs when the continuously compounded return is above zero; a loss occurs when it is below zero. - Normality alone does not make those probabilities equal: that would require the return distribution to be centered at zero. Constant variance across all horizons - A normal model for continuously compounded returns does not imply that return variance stays constant as the horizon changes. - Under standard independent-increment assumptions, return variance scales with the length of the horizon.
4. A bank quotes a 6.00% nominal annual rate compounded monthly. The calculated value of the effective annual rate is closest to:
Answer: B. 6.17%
6.17% - Monthly compounding makes the monthly rate 0.06/12 = 0.005. - The effective annual rate is (1.005)^12 − 1 = 6.1678%, which rounds to 6.17%. 6.00% - 6.00% is the quoted nominal annual rate, not the effective annual rate. - Reporting it without applying the 12 monthly compounding periods omits the interest earned on interest. 6.18% - Using continuous compounding instead gives e^0.06 − 1 ≈ 6.1837%, which rounds to 6.18%. - The bank specifies monthly, not continuous, compounding.
5. A portfolio's 64 monthly active returns have a mean of 1.20% and standard deviation of 4.80%. For H0: μ ≤ 0 versus H1: μ > 0, the 5% critical t-value is 1.669. The analyst should most likely:
Answer: C. Reject the null and conclude the mean active return is positive
Reject the null and conclude the mean active return is positive - The standard error is 4.80%/√64 = 0.60%, giving a t-statistic of 1.20%/0.60% = 2.00. - Because 2.00 exceeds the one-sided 5% critical value of 1.669, reject H0 in favor of a positive mean active return. Fail to reject the null because 1.20% is economically small - Whether 1.20% per month is economically meaningful is separate from the hypothesis test. - The decision is to reject H0 because the t-statistic of 2.00 exceeds 1.669. Fail to reject the null because the sample mean is below 1.669 - The 1.669 critical value applies to a t-statistic, not to a sample mean expressed as a percentage. - Dividing 1.20% by its 0.60% standard error gives 2.00, which exceeds 1.669.
6. An investor deposits USD 10,000 at the beginning of each year for four years into an account earning 6% annually. The calculated value at the end of Year 4 is closest to:
Answer: B. USD 46,371
USD 46,371 - Deposits at the beginning of each year form a four-payment annuity due. - Its end-of-Year-4 value is USD 10,000 × [(1.06⁴ − 1) / 0.06] × 1.06 = USD 46,370.93, closest to USD 46,371. USD 43,746 - USD 10,000 × [(1.06⁴ − 1) / 0.06] = USD 43,746.16. - This treats the deposits as an ordinary annuity made at each year-end, rather than at each year’s beginning. USD 49,153 - Compounding the annuity-due value for one extra year gives USD 46,370.9296 × 1.06 = USD 49,153.19. - That places the value at the end of Year 5, not the requested end of Year 4.
7. A stock follows geometric Brownian motion with annual drift 8%, volatility 20%, and simulated standard-normal shock 0.50. The one-year continuously compounded return is closest to:
Answer: C. 16.00%
16.00% - Under geometric Brownian motion, the one-year continuously compounded return is \((\mu-\tfrac12\sigma^2)t+\sigma\sqrt{t}Z\). - Substituting the given values yields \((0.08-\tfrac12(0.20)^2)(1)+0.20(0.50)=\mathbf{0.16}\), or 16.00%. 6.00% - The drift after the volatility adjustment is \(0.08-\tfrac12(0.20)^2=\mathbf{0.06}\). - This answer omits the simulated shock of \(0.20(0.50)=0.10\). 11.00% - This answer starts with the adjusted drift of 0.06 but adds only half the shock contribution: \(0.06+0.05=0.11\). - The full shock contribution is \(0.20(0.50)=0.10\), giving 0.16 instead.
8. An analyst models a security's continuously compounded one-period return as normally distributed. The security's ending price is most likely:
Answer: B. Lognormally distributed with strictly positive values
Lognormally distributed with strictly positive values - The continuously compounded return is \(r=\ln(P1/P0)\), so the ending price is \(P1=P0e^r\). - Because \(r\) is normal, exponentiating it makes \(P1\) lognormal and strictly positive for a positive beginning price. Normally distributed with possible negative values - The return, not the ending price, is modeled as normally distributed. - Treating the price as normal ignores \(P1=P0e^r\), which cannot produce a negative price when \(P0>0\). Uniformly distributed with a fixed upper bound - Exponentiating a normal return produces a lognormal, not a uniform, ending-price distribution. - A normal return has no fixed upper bound, so \(P0e^r\) has no fixed upper bound either.
9. An investor deposits USD 12,000 at each year-end for three years into an account earning 4% annually. The calculated value immediately after the third deposit is closest to:
Answer: B. USD 37,329
USD 37,329 - Because deposits occur at year-end, they earn two, one, and zero years of interest, respectively: USD 12,000(1.04² + 1.04 + 1) = USD 37,459.20. - USD 37,329 is the closest choice to that calculated value, although it is not the exact future value. USD 36,960 - USD 12,000(1.04 + 1.04 + 1) = USD 36,960. - This caps the first deposit’s growth at one year instead of the two years it earns before the third deposit. USD 38,958 - USD 37,459.20 × 1.04 = USD 38,957.568, or USD 38,958 rounded to the nearest dollar. - That extra year of growth treats the deposits as a beginning-of-year annuity, rather than year-end deposits valued immediately after the third deposit.
1. In Portfolio Management (Risk and Return), what is the formula that defines the correlation coefficient using covariance?
\[\rho{1,2} = \frac{Cov{1,2}}{\sigma1 \sigma2}\]
2. How is variance calculated from a probability model?
It is the probability-weighted sum of the squared deviations from the expected value (mean).
3. What is the arithmetic mean return?
The simple average of a series of periodic returns.
4. What does skewness measure in a distribution?
The extent to which a distribution is not symmetrical about its mean.
5. What is the possible range of values for a correlation coefficient?
From \(-1.0\) to \(+1.0\).
6. What is the median of a dataset?
The midpoint of a dataset when the data are arranged in ascending or descending order.
7. What is the mode of a dataset?
The value that occurs most frequently. (For continuous data, it is the modal interval).
8. What is the formula for the correlation coefficient \(\rho{XY}\)?
\[\rho{XY} = \frac{s{XY}}{sX sY}\] where \(s{XY}\) is covariance, and \(sX, sY\) are standard deviations.
9. What formula do you use to measure relative dispersion (risk per unit of expected return)?
The Coefficient of Variation (CV): \[CV = \frac{s}{\bar{X}}\]
Get full Oncourse access with lessons, practice questions, flashcards and AI study tools.
Scan to download app
Enter your email to get your 75% OFF code and unlock the full CFA question bank on the app.