You are reading the 2026 curriculum (exams up to November 2026). Sitting in February 2027 or later? Go to the 2027 Quantitative Methods guide.
Downside deviation and coefficient of variation
Target downside deviation measures spread only below a target return B: square the shortfalls, add them, divide by n − 1 (all observations), take the root. The coefficient of variation divides the standard deviation by the mean, giving risk per unit of return, so lower is better.
Before this: Range, mean absolute deviation, variance and standard deviation
On the exam
- Calculate, interpret, and evaluate measures of dispersion to address an investment problem
Calculate and interpret target downside deviation and the coefficient of variation. Expect the n − 1 divisor to be tested, a change of target, and a comparison of two funds by CV.
How deeply to learn each part
- Know the idea
- Understand it in general. It frames other topics, but you will not calculate with it.
- Use it
- Understand it well enough to solve a question with it. Recognize when it applies.
- Memorize
- Learn it by heart. The exam will not give it to you.
You know this one. The full lesson is folded away. The takeaways, formulas and checkpoint below are your quick refresher, and the checkpoint is a fair test of whether you still do.
You tested strong here. The full lesson is folded away. The takeaways, formulas and checkpoint below are your quick refresher, and the checkpoint is a fair test of whether you still know it.
How do you measure only the bad volatility?
The standard deviation counts a surprise gain exactly like a surprise loss. Most investors do not see it that way. Downside risk is the risk of returns falling below a level you care about, such as a minimum return needed to meet a liability.
The target downside deviation , also called the target semideviation , measures the spread of returns below a target B. Only observations below B enter the sum, but the divisor still counts every observation.
Of 8 returns, those below the target have squared shortfalls that add up to 19.21 (in %²). The target downside deviation is closest to:
√(19.21 ÷ 7) = √2.7443 = 1.66%
Target downside deviation1.66%
- 19.21÷(8−1)=√xn counts every observation, not only those below the target
√x square root
- 19.21ENTER8ENTER1−÷g√xn counts every observation, not only those below the target
ENTER enter (on the HP, puts the number on the stack) · g blue shift key · √x square root
Try this 0 of 2
- Set the target to 0% and check the downside deviation is 1.04%
- Raise the target to 3% or more and watch the downside deviation
Drag the dashed target line, or focus it and use the arrow keys. Orange months are below the target; the number under each is its squared shortfall.
4 months fall below 1.0%. Their squared shortfalls add to 19.21; divided by n − 1 = 7, the root is 1.66%, below the full standard deviation of 2.26%.
Show working
Shortfalls below B = 1.0%: Feb (−2.4 − 1.0)² = 11.56; Apr (0.5 − 1.0)² = 0.25; May (−1.2 − 1.0)² = 4.84; Aug (−0.6 − 1.0)² = 2.56.
Months at or above the target count as zero; they still count in n.
Target downside deviation = √(19.21 ÷ (8 − 1)) = √2.7443 = 1.66%.
Trap: dividing by the 4 months below target gives √(19.21 ÷ 4) = 2.19%.
Full sample standard deviation (mean 0.98%, all 8 months, ÷ 7) = 2.26%.
- 3.4x²+0.5x²+2.2x²+1.6x²=19.21, sum of squared shortfalls
- ÷7=2.7443
- √x1.66%
x² square · √x square root
- 3.4ENTER×0.5ENTER×+2.2ENTER×+1.6ENTER×+19.21, sum of squared shortfalls
- 7÷2.7443
- g√x1.66%
ENTER enter (on the HP, puts the number on the stack) · g blue shift key · √x square root
Drag the target line and watch the bars that turn orange. More bars fall short as the target rises, and each shortfall grows.
With a target near the average, the downside deviation is usually lower than the standard deviation, because only part of the data counts. Raise the target far enough and it can exceed it.
What is the coefficient of variation?
Standard deviations are hard to compare across assets with very different average returns. A 10% standard deviation is mild for an asset that earns 20% and wild for one that earns 2%. The coefficient of variation (CV) fixes this by expressing the standard deviation relative to the mean. It is a measure of relative dispersion .
A fund has a mean return of 9.00% and a standard deviation of 14.00%. Its coefficient of variation is closest to:
14.00% ÷ 9.00% = 1.5556
Coefficient of variation, CV1.5556
- 14÷9=both in percent, so the units cancel
- 14ENTER9÷both in percent, so the units cancel
ENTER enter (on the HP, puts the number on the stack)
- Standard deviation Memorize all spread around the mean, up and down alike
- Target downside deviation Memorize only shortfalls below a target B
- Divisor n − 1, with n counting every observation
- Higher target more observations count, bigger shortfalls, larger result
-
- Coefficient of variation Memorize standard deviation ÷ mean, a relative measure
- Reading it risk per unit of return, lower is better
- Caution only meaningful with a positive mean not near zero
-
Exam tip CV: lower is better. If a mean is zero or negative, do not use the CV at all.
Go deeper From downside deviation to the Sortino ratio
The Sortino ratio is the Sharpe ratio’s cousin: it divides the excess return over a target by the target downside deviation instead of by the standard deviation. It rewards a manager whose volatility comes mostly from gains. It appears in the portfolio management performance-measures material, and this lesson supplies its denominator.
Go deeper Why investors with a minimum return watch the downside
A pension fund that must earn at least 3% a year to pay benefits is not troubled by returns above 3%. Its pain comes from the months and years below that line. A manager with big upside swings and few shortfalls can have a high standard deviation but a low downside deviation, and the second number is the one that matches the fund’s problem.
Memorize this
MemorizeDownside deviation and CV
| Measure | Formula | Better when |
|---|---|---|
| Target downside deviation | √[ Σ(X − B)² for X < B ÷ (n − 1) ] | lower |
| Coefficient of variation | s ÷ X̄ | lower |
Memory hook Downside deviation: keep only the bad months, square the misses, but divide by all n − 1, not by how many were bad.
Know it cold Raise the target B and the downside deviation can only rise. CV is unit-free, so it compares assets with different returns.
- Standard deviation Memorize all spread around the mean, up and down alike
- Target downside deviation Memorize only shortfalls below a target B
- Divisor n − 1, with n counting every observation
- Higher target more observations count, bigger shortfalls, larger result
-
- Coefficient of variation Memorize standard deviation ÷ mean, a relative measure
- Reading it risk per unit of return, lower is better
- Caution only meaningful with a positive mean not near zero
-
Quick example
Returns 3%, −1%, 2%, −4%, 5% and a 0% target. Shortfalls −1 and −4: (1 + 16) ÷ (5 − 1) = 4.25, and √4.25 = 2.06%.
- 1x²+4x²= 17
- ÷4=√x 2.06
x² square · √x square root
- 1ENTER×4ENTER×+ 17
- 4÷g√x 2.06
ENTER enter (on the HP, puts the number on the stack) · g blue shift key · √x square root
Quick example
Mean 10%, standard deviation 12%. CV = 12 ÷ 10 = 1.20 units of risk per unit of return.
- 12÷10= 1.2
- 12ENTER10÷ 1.2
ENTER enter (on the HP, puts the number on the stack)
How it is asked
-
Target downside deviation from a short list
Returns 3%, −1%, 2%, −4%, 5%, target 0%. Target downside deviation?
(1 + 16) ÷ 4 = 4.25, √ = 2.06%. Traps: 2.92% (÷ 2 below-target values), 3.54% (full standard deviation).
-
Which fund has less risk per unit of return?
X: mean 6%, s 9%. Y: mean 10%, s 12%.
CV X = 1.50, CV Y = 1.20. Y wins. Trap: picking the lower standard deviation.
-
Effect of raising the target
The target return is raised. What happens to downside deviation?
It increases (shortfalls get bigger, more observations count).
Exam traps
- Dividing by the number of below-target observations instead of n − 1 (n counts every observation).
- Including above-target observations in the sum.
- Reading a higher CV as better. A lower CV means less risk per unit of return.
- Computing the CV with a mean near zero or negative, which makes it meaningless.
Checkpoint
4 exam-style questions. About 90 seconds each.
1Five annual returns are 3%, −1%, 2%, −4% and 5%. With a target of 0%, the target downside deviation is closest to:
2Fund X has a mean return of 6% and a standard deviation of 9%. Fund Y has a mean of 10% and a standard deviation of 12%. Which has less risk per unit of return?
3If the target return used in a downside deviation is raised, the downside deviation most likely:
4Ten monthly returns include three that fall below the target, with squared shortfalls that add up to 12.6 (%²). The target downside deviation is closest to:
Key takeaways
- Target downside deviation: shortfalls below B only, divided by n − 1.
- CV = s ÷ mean. Lower means less risk per unit of return.
- A higher target means a larger downside deviation.
- CV has no units, so it compares funds of different size and return.
Quick example
Returns 3%, −1%, 2%, −4%, 5% and a 0% target. Shortfalls −1 and −4: (1 + 16) ÷ (5 − 1) = 4.25, and √4.25 = 2.06%.
- 1x²+4x²= 17
- ÷4=√x 2.06
x² square · √x square root
- 1ENTER×4ENTER×+ 17
- 4÷g√x 2.06
ENTER enter (on the HP, puts the number on the stack) · g blue shift key · √x square root
Quick example
Mean 10%, standard deviation 12%. CV = 12 ÷ 10 = 1.20 units of risk per unit of return.
- 12÷10= 1.2
- 12ENTER10÷ 1.2
ENTER enter (on the HP, puts the number on the stack)
Questions about this lesson?
Where this comes back
- Portfolio Risk and Return: Part II (CAPM and Performance Measures) Portfolio Construction
- Skewness and kurtosis Statistical Measures of Asset Returns
Questions candidates ask
What is target semideviation?
Another name for target downside deviation. It measures the spread of returns below a chosen target B and ignores everything above it. Square each shortfall (X − B for X below B), add them, divide by n − 1 where n counts all observations, and take the square root.
How do you calculate the coefficient of variation?
Divide the standard deviation by the mean: CV = s ÷ X̄. A fund with a 9% standard deviation and a 7% mean has a CV of 1.29. It expresses risk per unit of return and has no units, which is why it can compare assets with different average returns.
Is a higher coefficient of variation better?
No. A higher CV means more risk for each unit of return, so a lower CV is better. It only makes sense when the mean is positive and not close to zero, because a tiny mean inflates the ratio.
Spotted an error? Tell us and we will fix it.