Why datetime handling matters in Pandas - Challenge Your Understanding

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this datetime conversion?

Consider this pandas code that converts a string column to datetime. What will be the output of the df['date'].dtype?

Pandas

import pandas as pd

df = pd.DataFrame({'date': ['2023-01-01', '2023-02-15', '2023-03-30']})
df['date'] = pd.to_datetime(df['date'])
print(df['date'].dtype)

A object

B datetime64[ns]

C string

D int64

💡 Hint

Think about what pandas does when converting strings to datetime format.

When pandas converts a string column to datetime using pd.to_datetime, the dtype changes from object (string) to datetime64[ns], which is pandas' internal datetime type.

❓ data_output

intermediate

How many rows are in this filtered datetime DataFrame?

Given this DataFrame with datetime values, how many rows remain after filtering for dates in March 2023?

Pandas

import pandas as pd

df = pd.DataFrame({'date': pd.to_datetime(['2023-01-10', '2023-03-05', '2023-03-20', '2023-04-01'])})
filtered = df[(df['date'] >= '2023-03-01') & (df['date'] < '2023-04-01')]
print(len(filtered))

A 1

B 4

C 3

D 2

💡 Hint

Count only dates that are in March 2023.

Only the dates '2023-03-05' and '2023-03-20' fall within March 2023, so the filtered DataFrame has 2 rows.

🔧 Debug

advanced

Why does this code raise an error when subtracting dates?

This code tries to subtract two date columns but raises an error. What is the cause?

Pandas

import pandas as pd

df = pd.DataFrame({'start': ['2023-01-01', '2023-02-01'], 'end': ['2023-01-10', '2023-02-15']})
df['duration'] = df['end'] - df['start']
print(df['duration'])

A TypeError because columns are strings, not datetime objects

B KeyError because 'duration' column does not exist

C ValueError because dates are in wrong format

D No error, outputs timedelta values

💡 Hint

Check the data types of the 'start' and 'end' columns before subtraction.

The columns 'start' and 'end' are strings, so subtracting them raises a TypeError. They must be converted to datetime first.

❓ visualization

advanced

Which plot shows correct monthly sales trend?

Given monthly sales data with datetime index, which plot code correctly shows the sales trend over time?

Pandas

import pandas as pd
import matplotlib.pyplot as plt

sales = pd.Series([100, 150, 200, 250], index=pd.to_datetime(['2023-01-01', '2023-02-01', '2023-03-01', '2023-04-01']))

A

sales.plot()
plt.show()

B

plt.plot(sales.index, sales.values)
plt.show()

C

plt.bar(sales.index, sales.values)
plt.show()

D

plt.scatter(sales.values, sales.index)
plt.show()

💡 Hint

Think about which plot type best shows a trend over time.

Using sales.plot() creates a line plot with datetime index on x-axis, showing the trend clearly. Bar or scatter plots are less suitable here.

🧠 Conceptual

expert

Why is timezone handling important in datetime data?

Which statement best explains why handling timezones correctly is crucial in datetime data analysis?

A Because timezone info is required to convert datetime to string

B Because timezones determine the date format used in the data

C Because ignoring timezones can cause incorrect time comparisons and data misalignment

D Because timezones affect the speed of datetime calculations

💡 Hint

Think about what happens if you compare times from different zones without adjustment.

Ignoring timezones can lead to wrong conclusions when comparing or merging datetime data from different regions, causing errors in analysis.