How is the interquartile range used to identify outliers in financial data?

Question

AcadiFi · Accepted Answer

The interquartile range (IQR) is a robust dispersion measure that focuses on the middle 50% of data, making it resistant to extreme values that can distort standard deviation.

## Calculation

**IQR = Q3 - Q1**

Where Q1 is the 25th percentile and Q3 is the 75th percentile.

## Outlier Detection — The 1.5 x IQR Rule

A common convention (also used on the CFA exam) defines outliers as:
- **Lower outlier:** any observation below Q1 - 1.5 x IQR
- **Upper outlier:** any observation above Q3 + 1.5 x IQR

These boundaries are sometimes called **fences**.

## Worked Example — Northgate Fund Monthly Returns

12 months of returns for Northgate Global Equity Fund (sorted):

-8.2%, -3.1%, -1.5%, 0.4%, 1.2%, 1.8%, 2.3%, 2.9%, 3.6%, 4.1%, 5.5%, 14.7%

**Step 1:** Q1 = average of 3rd and 4th values = (-1.5 + 0.4) / 2 = **-0.55%**

**Step 2:** Q3 = average of 9th and 10th values = (3.6 + 4.1) / 2 = **3.85%**

**Step 3:** IQR = 3.85 - (-0.55) = **4.40%**

**Step 4:** Fences:
- Lower fence = -0.55 - 1.5(4.40) = **-7.15%**
- Upper fence = 3.85 + 1.5(4.40) = **10.45%**

**Results:**
- -8.2% falls below -7.15% → **outlier**
- 14.7% falls above 10.45% → **outlier**

## Why IQR Matters in Finance

Standard deviation is sensitive to outliers — one extreme month can inflate σ by 30-40%. This matters when:
- Screening hedge fund returns for potential data errors
- Building factor models where outlier observations can dominate regressions
- Reporting risk metrics to clients who want a "typical" range of returns

## IQR vs Standard Deviation

| Feature | IQR | Standard Deviation |
|---|---|---|
| Outlier sensitivity | Resistant | Highly sensitive |
| Uses all data points | No (only middle 50%) | Yes |
| Assumes normality | No | Implicitly |
| Best for | Skewed or heavy-tailed data | Normal or near-normal data |

**Exam tip:** CFA Level I may give you a sorted dataset and ask you to compute IQR or identify outliers. Practice the percentile calculation — it is easy points.

Practice more in our CFA Level I question bank.

How is the interquartile range used to identify outliers in financial data?

Master Level I with our CFA Course

Related Questions

Practice Questions