Outlier detection with IQR in Pandas - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this IQR outlier detection code?

Given the DataFrame df below, what does the outliers variable contain after running the code?

Pandas

import pandas as pd

df = pd.DataFrame({'values': [10, 12, 14, 15, 18, 19, 20, 100]})
Q1 = df['values'].quantile(0.25)
Q3 = df['values'].quantile(0.75)
IQR = Q3 - Q1
outliers = df[(df['values'] < Q1 - 1.5 * IQR) | (df['values'] > Q3 + 1.5 * IQR)]
print(outliers)

A

   values
7     100

B Empty DataFrame\nColumns: [values]\nIndex: []

C

   values
0      10
7    100

D

   values
0     10

💡 Hint

Calculate Q1 and Q3, then find the IQR. Check which values fall outside the range Q1 - 1.5*IQR to Q3 + 1.5*IQR.

The IQR is Q3 - Q1 = 19 - 12 = 7. The lower bound is 12 - 1.5*7 = 1.5 and the upper bound is 19 + 1.5*7 = 29.5. Only the value 100 is above 29.5, so it is the only outlier.

❓ data_output

intermediate

How many outliers are detected by this IQR method?

Using the DataFrame df and the IQR method below, how many rows are identified as outliers?

Pandas

import pandas as pd

df = pd.DataFrame({'scores': [55, 60, 65, 70, 75, 80, 85, 90, 95, 200]})
Q1 = df['scores'].quantile(0.25)
Q3 = df['scores'].quantile(0.75)
IQR = Q3 - Q1
outliers = df[(df['scores'] < Q1 - 1.5 * IQR) | (df['scores'] > Q3 + 1.5 * IQR)]
print(len(outliers))

A 2

B 0

C 1

D 3

💡 Hint

Calculate the IQR and check which values fall outside the bounds.

Q1 is 67.5, Q3 is 87.5, so IQR is 20. Lower bound is 67.5 - 1.5*20 = 37.5, upper bound is 87.5 + 1.5*20 = 117.5. Only 200 is outside this range, so 1 outlier.

❓ visualization

advanced

Which boxplot correctly shows the outliers detected by IQR?

You run this code to detect outliers and plot a boxplot. Which option shows the correct boxplot visualization?

Pandas

import pandas as pd
import matplotlib.pyplot as plt

df = pd.DataFrame({'data': [5, 7, 8, 9, 10, 12, 15, 100]})
plt.boxplot(df['data'])
plt.show()

A Boxplot with one point far below the lower whisker at 5

B Boxplot with one point far above the upper whisker at 100

C Boxplot with multiple points above whiskers at 15 and 100

D Boxplot with no points outside whiskers

💡 Hint

Outliers appear as points outside the whiskers. The value 100 is much larger than others.

The boxplot shows one outlier point at 100 above the upper whisker. The lower whisker includes 5, so no outlier below.

🔧 Debug

advanced

What error does this IQR outlier detection code raise?

Identify the error raised by this code snippet:

Pandas

import pandas as pd

df = pd.DataFrame({'vals': [1, 2, 3, 4, 5]})
Q1 = df['vals'].quantile(0.25)
Q3 = df['vals'].quantile(0.75)
IQR = Q3 - Q1
outliers = df[(df['vals'] < Q1 - 1.5 * IQR) | (df['vals'] > Q3 + 1.5 * IQR)]
print(outlier)

A SyntaxError: invalid syntax

B TypeError: unsupported operand type(s) for -: 'str' and 'float'

C KeyError: 'vals'

D NameError: name 'outlier' is not defined

💡 Hint

Check variable names carefully in the print statement.

The variable is named 'outliers' but the code tries to print 'outlier', which is undefined.

🚀 Application

expert

Which option correctly filters out outliers using IQR in a DataFrame with multiple columns?

You have a DataFrame df with columns A and B. You want to remove rows where A or B have outliers based on IQR. Which code correctly does this?

Pandas

import pandas as pd

df = pd.DataFrame({'A': [10, 12, 14, 100, 15], 'B': [20, 22, 23, 24, 200]})

A

Q1 = df.quantile(0.25)
Q3 = df.quantile(0.75)
IQR = Q3 - Q1
filtered = df[(df['A'] >= Q1['A'] - 1.5 * IQR['A']) & (df['A'] <= Q3['A'] + 1.5 * IQR['A']) & (df['B'] >= Q1['B'] - 1.5 * IQR['B']) & (df['B'] <= Q3['B'] + 1.5 * IQR['B'])]

B

Q1 = df.quantile(0.25)
Q3 = df.quantile(0.75)
IQR = Q3 - Q1
filtered = df[(df >= Q1 - 1.5 * IQR) & (df <= Q3 + 1.5 * IQR)]

C

Q1_A = df['A'].quantile(0.25)
Q3_A = df['A'].quantile(0.75)
IQR_A = Q3_A - Q1_A
Q1_B = df['B'].quantile(0.25)
Q3_B = df['B'].quantile(0.75)
IQR_B = Q3_B - Q1_B
filtered = df[~((df['A'] < Q1_A - 1.5 * IQR_A) | (df['A'] > Q3_A + 1.5 * IQR_A) | (df['B'] < Q1_B - 1.5 * IQR_B) | (df['B'] > Q3_B + 1.5 * IQR_B))]

D

Q1 = df.quantile(0.25)
Q3 = df.quantile(0.75)
IQR = Q3 - Q1
filtered = df[(df['A'] >= Q1['A'] - 1.5 * IQR['A']) | (df['A'] <= Q3['A'] + 1.5 * IQR['A']) | (df['B'] >= Q1['B'] - 1.5 * IQR['B']) | (df['B'] <= Q3['B'] + 1.5 * IQR['B'])]

💡 Hint

Use logical AND to keep rows where both columns are within the IQR bounds.

Option A correctly applies the IQR filter for each column using AND conditions to keep rows where neither column is an outlier. Option A uses negation but is correct logically; however, it is verbose. Option A uses DataFrame-wide comparison which returns NaNs and does not filter properly. Option A uses OR with >= and <= which includes all rows and is incorrect.