Resampling time series in Data Analysis Python - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of resampling with mean aggregation

What is the output of this code snippet that resamples a time series to daily frequency and calculates the mean?

Data Analysis Python

import pandas as pd
import numpy as np

dates = pd.date_range('2024-01-01 12:00', periods=4, freq='6H')
data = pd.Series([10, 20, 30, 40], index=dates)
result = data.resample('D').mean()
print(result)

A

2024-01-01    15.0
2024-01-02    35.0
dtype: float64

B

2024-01-01    10.0
2024-01-02    20.0
dtype: float64

C

2024-01-01    25.0
2024-01-02    35.0
dtype: float64

D

2024-01-01    20.0
2024-01-02    40.0
dtype: float64

💡 Hint

Think about how resample groups data by day and then averages values within each day.

The data has 4 values at 6-hour intervals starting at noon on Jan 1. Resampling by day groups values on Jan 1 (12:00 and 18:00) and Jan 2 (00:00 and 06:00). The mean for Jan 1 is (10+20)/2=15, for Jan 2 is (30+40)/2=35.

❓ data_output

intermediate

Number of rows after upsampling with forward fill

Given a time series with hourly data for 3 hours, what is the number of rows after upsampling to 30-minute frequency with forward fill?

Data Analysis Python

import pandas as pd

dates = pd.date_range('2024-01-01 00:00', periods=3, freq='H')
data = pd.Series([1, 2, 3], index=dates)
upsampled = data.resample('30T').ffill()
print(len(upsampled))

A 7

B 5

C 4

D 6

💡 Hint

Count how many 30-minute intervals fit between the first and last timestamp inclusive.

Original data has 3 points at 00:00, 01:00, 02:00. Upsampling to 30 minutes creates timestamps at 00:00, 00:30, 01:00, 01:30, 02:00. That's 5 rows. Forward fill fills missing values.

🔧 Debug

advanced

Identify the error in resampling code

What error does this code raise when trying to resample a DataFrame without a datetime index?

Data Analysis Python

import pandas as pd

data = pd.DataFrame({'value': [1, 2, 3]})
result = data.resample('D').sum()

A TypeError: Only valid with DatetimeIndex, TimedeltaIndex or PeriodIndex, but got an instance of 'RangeIndex'

B KeyError: 'D'

C AttributeError: 'DataFrame' object has no attribute 'resample'

D ValueError: Invalid frequency string

💡 Hint

Check the type of index required for resample to work.

Resample requires the DataFrame to have a datetime-like index. Here, the index is default RangeIndex, so it raises a TypeError indicating the index type is invalid.

🚀 Application

advanced

Choosing correct resampling method for downsampling

You have minute-level temperature data and want to get the maximum temperature per hour. Which resampling method and aggregation should you use?

A resample('D').sum()

B resample('T').mean()

C resample('H').max()

D resample('H').min()

💡 Hint

Think about grouping data into hours and picking the highest temperature.

To get max temperature per hour, resample with hourly frequency ('H') and aggregate with max(). Other options either use wrong frequency or wrong aggregation.

🧠 Conceptual

expert

Effect of resampling with different label and closed parameters

When resampling time series data with 'label="right"' and 'closed="right"', how are the bins labeled and which side is included in the interval?

A Bins are labeled by the left edge and intervals include the left edge but exclude the right edge

B Bins are labeled by the left edge and intervals include both edges

C Bins are labeled by the right edge and intervals include the left edge but exclude the right edge

D Bins are labeled by the right edge and intervals include the right edge but exclude the left edge

💡 Hint

Think about how intervals are defined when closed='right' and how labels align.

With label='right' and closed='right', each bin is labeled by its right endpoint, and intervals include the right edge but exclude the left edge. This affects which data points fall into each bin.